This is a reproduction of a library book that was digitized
by Google as part of an ongoing effort to preserve the
information in books and make it universally accessible.
Google" books
https://books.google.com
Digitized by
Google
Digitized by v^ooQle
Digitized
i by Google
MONOGRAPHS ON BIOCHEMISTRY.
Edited by R. H. A. Plimmer, D.Sc., and P. G. Hopkins, F.R.S., D.Sc.
Royal 8vo.
THE NATURE OF ENZYME ACTION. By W. M. Bayliss,
M.A., D.Sc., F.R.S , Professor of General Physiology, University College, London.
5J. net.
THE CHEMICAL CONSTITUTION OF THE PROTEINS.
By R. H. A. Plimmer, D.Sc. Two Parts.
Part I., Analysis. 5*. 6 d. net.
Part II., Synthesis, etc. y. 6d. net.
THE GENERAL CHARACTERS OF THE PROTEINS. By
S. B. Schryver, Ph.D., D.Sc., Lecturer on Physiological Chemistry, University
College, London. 2s. 6d. net.
THE VEGETABLE PROTEINS. By Thomas B. Osborne, Ph.D.,
Research Chemist in the Connecticut Agricultural Experiment Station, New Haven.
Connecticut; Research Associate of the Carnegie Institution of Washington, D.C.
y. 6d. net
THE SIMPLE CARBOHYDRATES AND THE GLUCOSIDES.
By E. Frankland Armstrong, D.Sc., Ph.D.* Fellow of the City and Guilds of
London Institute, y. net.
THE FATS. By J. B. Leathes, F.R.S., M.A., M.B., F.R.C.S.,
Professor of Pathological Chemistry in the University of Toronto. 4s. net.
ALCOHOLIC FERMENTATION. By A. Harden, Ph.D., D.Sc.,
F.R.S., Professor of Biochemistry, London University. 4s. net.
THE PHYSIOLOGY OF PROTEIN METABOLISM. By E. P.
Cathcart, M.D., D.Sc., of Glasgow. Ch.B., Professor of Physiology at the Lon¬
don Hospital Medical College, University of London. 4s. 6 d. net.
SOIL CONDITIONS AND PLANT GROWTH. By Edward J.
Russell, D.Sc. (Lond.), Goldsmiths’ Company’s Soil Chemist, Rothamsted Ex¬
perimental Station, Harpenden. With Diagrams. 5J. net.
OXIDATIONS AND REDUCTIONS IN THE ANIMAL BODY.
By H. D. Dakin, D.Sc., F.I.C., The Herter Laboratory, New York. 4s. net.
THE SIMPLER NATURAL BASES. By George Barger, M.A.,
D.Sc., Professor of Chemistry in the Royal Holloway College. 6 j. net.
NUCLEIC ACIDS. Their Chemical Properties and Physiological
Conduct. By Walter Jones, Ph.D., Professor of Physiological Chemistry in the
Johns Hopkins Medical School, y . 6d. net.
THE RESPIRATORY EXCHANGE OF ANIMALS AND MAN.
By August Krogh, Ph.D., Professor of Zoophysiology in the University of
Copenhagen.
The following are in preparation :—
THE DEVELOPMENT AND PRESENT POSITION OF BIO¬
LOGICAL CHEMISTRY. By F. Gowland Hopkins, D.Sc., F.R.S.,
Professor of Biochemistry in the University of Cambridge.
THE POLYSACCHARIDES. By Arthur R. Ling, F.I.C.
COLLOIDS. By W. B. Hardy, M.A., F.R.S.
PROTAMINES AND HISTONES. By A. Kossel, Ph.D.
ORGANIC COMPOUNDS OF ARSENIC AND ANTIMONY.
By Gh.bk.rt T. Morgan, D.Sc., F.I.C.
LECITHIN AND ALLIED SUBSTANCES. By Hugh Mac-
lean, M.D., D.Sc.
THE ORNAMENTAL PLANT PIGMENTS. By A. G. Perkin,
F R S
CHLOROPHYLL AND HAEMOGLOBIN. By H. J. Page, B.Sc.
LONGMANS, GREEN AND CO.
LONDON, NEW YORK, BOMBAY, CALCUTTA, AND MADRAS.
Digitized by v^ooQle
MONOGRAPHS ON PHYSICS.
Edited by Sir J. J. Thomson, O.M., F.R.S., Cavendish Professor of Experimental
Physics, Cambridge, and Frank Horton, Sc.D., Professor of Physics in the
University of London.
8vo.
RAYS OF POSITIVE ELECTRICITY AND THEIR APPLICA¬
TION TO CHEMICAL ANALYSIS. By Sir J. J. Thomson, O.M., F.R.S.,
Cavendish Professor of Experimental Physics, Cambridge, and Professor of Natural
Philosophy at the Royal Institution, London. With Illustrations. 5-r. net.
MODERN SEISMOLOGY. By G. W. Walker, A.R.C.Sc., M.A.,
F.R.S., formerly Fellow of Trinity College, Cambridge. With Plates and Diagrams.
y. net.
PHOTO-ELECTRICITY, THE LIBERATION OF ELECTRONS
BY LIGHT; with Chapters on Fluorescence and Phosphorescence, and Photo-
Chemical Actions and Photography. By H. Stanley Allen, M.A., D.Sc., Senior
Lecturer on Physics at University of London, King's College. With Diagrams.
7-f. 6 d. net.
THE SPECTROSCOPY OF THE EXTREME ULTRA-VIOLET.
By Theodore Lyman, Ph.D., Assistant Professor of Physics, Harvard University.
With Diagrams. 5?. net.
THE PHYSICAL PROPERTIES OF COLLOIDAL SOLUTIONS.
By E. F. Burton, B.A., Ph.D., Associate Professor of Physics, The University,
Toronto.
RELATIVITY AND THE ELECTRON THEORY. By E.
Cunningham, M.A., Fellow and Lecturer of St. John's College, Cambridge.
ELECTRIC WAVES. By G. W. Pierce, Professor of Physics,
Harvard University. [In preparation .
THE EMISSION OF ELECTRICITY FROM HOT BODIES. By
O. W. Richardson, F.R.S., Wheatstone Professor of Physics, King's College,
London. [In preparation .
ATMOSPHERIC IONIZATION. By J. C. McLennan, F.R.S.,
Professor of Physics, The University, Toronto. [In preparation .
LONGMANS, GREEN AND CO.
LONDON, NEW YORK, BOMBAY, CALCUTTA, AND MADRAS.
Digitized by v^ooQle
Monographs on Inorganic and
Physical Chemistry
EDITED BY
ALEXANDER FINDLAY, M.A., D.Sc., F.I.C.
Professor of Chemistry, University College of Wales, Aberystwyth.
To those engaged in guiding the reading of advanced
students of chemistry, the difficulty of obtaining ade¬
quately summarised accounts of the progress made in
recent years, more especially along certain of the more
actively pursued lines of advance, becomes ever more
acutely felt. So great has now become the volume of
chemical investigation, and so numerous the channels of
its publication, that not only the Honours Student but
also the worker desirous of undertaking Research in one
or other department of his subject, feels* it a growing
difficulty to become au fait with the present state of the
more important and more strenuously cultivated regions
of his Science. To assist these two classes of students—
those reading for an Honours Degree, and those under¬
taking Research—is the main aim of the present Series
of Monographs.
In this Series of Monographs it is hoped to place
before advanced students of Chemistry, accounts of certain
sections of Inorganic and Physical Chemistry fuller and
more extended in scope than can be obtained in ordinary
text-books. Exhaustive treatment of the different sub¬
jects, however, so far as concerns work important in its
time but now only of historical interest, will not be
attempted; the chief attention will be given to recent
investigations.
Digitized by CnOOQle
1
Monographs on Inorganic and Physical
Chemistry— {continued}
The following are now ready .—
THE CHEMISTRY OF THE RADIO-ELEMENTS. By
Frederick Soddy, F.R.S., Professor of Chemistry in the
University of Aberdeen.
Part I. 4*. net.
Part II. The Radio-Elements and the Periodic Law. 2 s. net.
Parts I. and II. in One Volume. 5*. 6 d. net.
PER-ACIDS AND THEIR SALTS. By T. Slater Price, D.Sc.,
of the Municipal Technical School, Birmingham. 3 s. net.
OSMOTIC PRESSURE. By Alexander Findlay, M.A., Ph.D.,
D.Sc., Professor of Chemistry and Director of the Edward Davies
Chemical Laboratories, University College of Wales. 2 s. 6 d. net.
INTERMETALLIC COMPOUNDS. By Cecil H. Desch, D.Sc.,
Graham Young Lecturer in Metallurgical Chemistry in the
University of Glasgow. With 17 Figures. 3*. net.
THE VISCOSITY OF LIQUIDS. By Albert Ernest Dunstan,
D.Sc. (Lond.), Head of the Chemical Department, East Ham
Technical College, and Ferdinand Bernard Thole, D.Sc.
(Lond.), Lecturer on Organic Chemistry, East Ham Technical I
College. 3*. net.
MOLECULAR ASSOCIATION. By W. E. S. Turner, D.Sc., M.Sc.,
Lecturer in Physical and Inorganic Chemistry in the University
of Sheffield. With 6 Figures. 5*. net.
THE MOLECULAR VOLUMES OF LIQUID CHEMICAL I
COMPOUNDS, from the Point of View of Kopp, By
Gervaise Le Bas, B.Sc. (Lond.). With Diagrams, 7 s. 6 d. net.
LONGMANS, GREEN, & CO., 39 Paternoster Row, London, E.C,
New York, Bombay, Calcutta, and Madras.
Digitized by
Google
MONOGRAPHS ON INORGANIC AND PHYSICAL CHEMISTRY
Edited by ALEXANDER FINDLAY, D.Sc.
THE MOLECULAR VOLUMES OF
LIQUID CHEMICAL COMPOUNDS
FROM THE POINT OF VIEW OF KOPP.
Digitized by C.oooLe
Digitized by
Google
THE
MOLECULAR VOLUMES
OF LIQUID
CHEMICAL COMPOUNDS
FROM THE POINT OF VIEW OF KOPP.
BY
GERVAISE LE BAS, B.Sc. (Lond.)
M
WITH DIAGRAMS
LONGMANS, GREEN AND CO.
39 PATERNOSTER ROW, LONDON
FOURTH AVENUE & 30th STREET, NEW YORK
BOMBAY, CALCUTTA, AND MADRAS
1915
[All rights reserved]
Digitized by Lnooole
Digitized by
Google
PREFACE.
The work which has been attempted in the present
volume is an application of the theory of molecular volumes
from the point of view of Kopp, to a study of the
constitutions of organic compounds chiefly. The working
out of the theory thus reverts to the line of its historic
development, the interval, representing a period of twenty-
five years, marking no advance at all along this line.
One reason is, that when the last important work was
done, the question of the possible influence of constitutive
influences on a physical property was not so prominent as
it is to-day. The writers were, for the most part, content
to point out that Kopp’s additive rule suffered limitations,
without inquiring into the reasons for this, and giving to
the deviations their proper interpretation. We might,
perhaps, except the work of Thorpe, which is thoroughly
imbued with the modern spirit. It is, however, just these
limitations of the additive rule which are so useful to the
chemist in his endeavour to work out the details of the
structure of chemical compounds. This has been turned
to good account in most of the physical properties, and the
success which has attended the work in these directions,
has probably contributed to the comparative neglect of
molecular volumes. Another reason is that, so long as
Kopp’s original values for carbon and hydrogen were
retained, it was impossible to put the subject on a satis¬
factory basis, because, under the conditions, the evidence
for one of the most important constitutive effects—that of
ring structure—was completely obliterated. The following
features of the old theory are retained: (a) Kopp’s con*
V
331002
Digitized by
Google
VI
PREFACE.
ception of molecular and atomic volumes, and (b) the
recognition of the additive principle. We are also
indebted to those careful and patient workers, through
whose laborious experimental researches that large amount
of data which at present exists has accumulated. This has
resulted in the detection of certain volume disturbances in
the case of particular compounds the meaning of which
was not at the time fully realized. These have of course
been considered, and given what we believe to be their
proper interpretation. Nearly all the atomic volumes
have been modified; some are strikingly different from
the old numbers. We also find it necessary to consider
a larger number of atomic values for say, oxygen or
nitrogen, than on the old theory. The result of these
changes in the atomic values is a corresponding change in
the conclusions drawn from the data. Whilst abundant
material then existed for a fairly extended theory, it was
found impossible to deal adequately with the subject,
chiefly because of the initial errors in the values of the
fundamental atoms, as well as on account of the im¬
perfect conceptions of chemical constitution which then
existed. We cannot claim to have elaborated a perfectly
satisfactory system, but the initial mistakes have been
corrected, and we believe that the work is progressing
on the right lines. At least there has been an honest
endeavour to correctly interpret facts, and we have not
knowingly subjected the evidence to an undue strain.
The difficulties which one is likely to meet with in
a research like the present one are great, for, owing to an
embarras des richesses , one cannot at once distinctly perceive
the relative importance of the facts which emerge, nor
clearly distinguish real from imaginary regularities. Facts
are liable to interpretation in so many ways. The want
of an unifying principle has been keenly felt, but we
believe that certain of the explanations which one feels
instinctively to be slightly unsatisfactory, and which do not
seem to fit in well with other explanations, are really
Digitized by v^ooQle
PkEFACE.
vii
different aspects of one and the same thing. There are
doubtless fundamental features which have not yet been
grasped, and which might serve to link up facts at
present somewhat disconnected.
Although we may have to wait for a thorough and
scientifically worked out theory, we desire to lay stress on
the good results which have already been obtained. The
chief of these is undoubtedly the influence of ring structure
and the evidence therefor. No physical property is ap¬
parently so well adapted to elucidate the ring structure of
a compound as molecular volumes, and it is to be hoped that
this will henceforward take its legitimate place among
other physical properties as an instrument of research.
A study of molecular volumes, especially, has shown
us that probably we are not at the end of the utility of
physical properties as means for giving us an insight into
the structure of molecules. When a more scientific
method of examination of these physical properties shall
have been worked out, there is no doubt that a great
advance will be made in a knowledge of the intricacies of
chemical constitution. It is doubtful if the possibilities in
this direction are as yet generally realized.
Closely connected with the above subject, are a number
of very important theoretical questions, such as the intimate
structure of liquids, the nature of the atom and molecule,
etc., but we have carefully avoided introducing these
speculative questions into a more or less systematic work
like the present
We might, perhaps, note in passing, that the theory of
molecular volumes, which is based upon an alternative
conception of the structure of liquids, viz.: that due to
Traube, has not been so successful as anticipated. An
examination of current work shows how pessimistic is the
spirit which obtains with reference to it Thus we find in
a recent textbook* on physical chemistry, “Beyond the
* The Relations between Chemical Constitution and some Physical
Properties , by Dr. Smiles, chap, iv., pp. 145-6.
Digitized by v^ooQle
PREFACE.
viii
detection of association, volume relations are of little use
in solving questions of constitution,” and again, .
The prospect of such a method is not very bright . . . . ”
Leaving aside the fact that Traube’s theory is based upon
a number of assumptions which are doubtful, we should
note that it does not answer to the test of all theory, viz.
utility as an instrument of research, and an accumulation
of good results obtained. A theory, like a tree, should be
estimated by the quantity and quality of its fruit. Hitherto
the crop has not been extensive.
The data have been chiefly obtained from Clarke’s
Constants of Nature , Part I., the Chemiker-Kalender ; and
from many of the original papers to be found in the
English and German scientific journals.
We desire to express our gratitude to Prof. W. J. Pope,
F.R.S., for the kindly help and encouragement extended
for a period of eight years or so, during which most of the
matter included in the present volume has been in course
of preparation.
G- Le B.
15 October , 1915.
Digitized by CnOOQle
CONTENTS.
CHAPTER PAGE
I. The Molecular Volumes of Organic Compounds at the
Melting-Point .i
The hydrocarbons : paraffins, olefins, and acetylenes. The
atomic volumes of carbon and hydrogen, i. The
influence of the alternating factor, 4. The effect of
unsaturation at the melting-point, 6. The nitriles,
7. Oxygen compounds : the alcohols, ketones, and
fatty acids. The volume of hydroxylic oxygen, 8.
The volume of ketonic oxygen, 9. The volume of
carboxylic oxygen, 11. Ring compounds, 14.
II. The Molecular Volumes of the Hydrocarbons at the
Boiling-Point.17
The open-chain compounds: paraffins, olefins, and di¬
olefins, 16. The volumes of carbon and oxygen, 17.
The influence of the homologous increment CH 2 , 22.
The branching of the hydrocarbon chain, 25I The
interaction between the iso group and ethenoid
linkage, 26. Closed-chain hydrocarbons—simple and
compound rings, condensed and separated rings, 33.
The effect of self-affinity on the side-chains of
aromatic compounds, 35. The polymethylenes, 46.
The naphthenes, 51. A general study of the
terpenes: hemi-terpenes, olefin terpenes, menthan
terpenes, camphan terpenes, sesqui-terpenes, and di-
terpenes, 53. The effect of cross-linking and bridging
of the ring on the magnitude of the contraction, 67.
Connexion with unsaturation, 68. Ring systems, 71.
ix
Digitized by v^ooQle
X
CONTENTS.
PAGE
III. The Molecular Volumes of the Halogen Compounds. 74
Fluorine, 74. Chlorine, 76. Bromine, 79. Iodine, 81.
The branching of the hydrocarbon chain, 83. Partial
rings, a, ft and y halogen compounds, 85. The in¬
fluence of the ethenoid linkage, 88. Application of
principles to complex compounds, 92. Aromatic
halogen compounds, 93. The gradual chlorination of
benzene, 94.
IV. The Molecular Volumes of Organic Compounds contain¬
ing Oxygen.97
The volume of oxygen in the alcohols, 98. The branched-
chain compounds, 99. The volumes of ft and y
di-hydroxy compounds, 100. The phenols, 104. Ter-
penic or ring alcohols, 106. The ethers, 108. The
position and nature of liquid water, 109. The effect of
the addition of the homologous increment CH 2 on the
volumes of the ethers, 113. The symmetrical and un-
symmetrical ethers, 116. Formula representing results
for complexity and symmetry in the ethers, 119. An
examination of the various constitutive effects associ¬
ated with hydrocarbon chains, 120. Complexity, 120.
Compactness of molecules, 121. Heterogeneity, 123.
Symmetry, 126. The phenolic and other ethers, 127.
The o, m, and p cresylic ethers, 130. Oxygen with
double linking, 133. The aldehydes and ketones, 134.
The carboxylic acids, 140. The fatty esters, 142. The
effect of the homologous increment on the ester series,
144. The methyl, ethyl, and propyl salts of the fatty
acids, 147. Causes of variation in volumes of isomers
and of the influence of symmetry, 150. Unsaturated
oxygen compounds, 153. The substituted esters,
155. The volume of oxygen (: 0 ) in union with sul¬
phur, nitrogen, and phosphorus, 161. The aromatic
acids and esters, 163. The volume of ring oxygen,
168. A summary of the mode of variation of the
oxygen atoms in organic compounds, 173.
V. The Molecular Volumes of Sulphur Compounds . .178
The volume of the element sulphur, 178. The halogen
derivatives of sulphur with and without sulphur, 180.
Digitized by
Google
CONTENTS.
xx
PAGE
A study of the constitutions of certain sulphur com¬
pounds investigated by Thorpe, 182. The sulphur
oxides, acids, and esters, 184. Doubly bound sulphur,
186. Ring compounds including sulphur, 187. The
thio-alcohols (mercaptans) and ethers, 188. Selenium
and compounds, 189. Chromium compounds, 190.
VI. The Molecular Volumes of Nitrogen Compounds . 191
The element nitrogen and NClj, 192. The nitriles and car-
bylamines, 193. Cyanic acid and the esters, 193.
Nitrogen compounds containing sulphur, 195. The
amines, 197. The diamines and derivatives including
ring compounds, 203. Aromatic amino compounds,
208. The oxides and esters of the nitrogen acids in¬
cluding the corresponding aromatic compounds, 213.
Alkaloid derivatives containing nitrogen, 222.
VII. The Molecular Volumes of Phosphorus Compounds, etc. 223
The element phosphorus and phosphorus oxides, 223.
Phosphorus compounds containing oxygen, halogens,
and sulphur, 227. Trivalent and pentavalent nitrogen,
229. Compounds of arsenic and antimony, 232.
Elements of Group 4: Silicon, titanium, vanadium, ger¬
manium, 233.
Miscellaneous elements: Boron, zinc, 234.
VIII. Summary of the Theory of Molecular Volumes . . 235
A. The Additive principle with Table of atomic volumes,
235. Variations in the volumes of the atoms, 237.
B. Constitutive influences: (a) The influence of the homo¬
logous increment, 239. ( 6 ) The influence of unsatura¬
tion, 241. (c) Partial ring structure, 245. (d) Ring
structure, 246. (e) Molecular volume and valency,
247. (/) The constitutive influences due to special
groups, 249.
C. A discussion of the special conception embodied in
molecular volumes, and the relation which obtains be¬
tween this property and other physical properties such
as boiling-point, surface tension, viscosity, etc., 249.
Digitized by v^ooQle
i CONTENTS .
PAGE
IX. Appendices:—
i. A Formula by means of which the molecular volume at
the boiling point may be calculated . . . .263
11. An Investigation of the Dicarboxylic Esters . . . 265
Table of References .273
Bibliography .275
Digitized by v^ooQle
CHAPTER I.
THE VOLUMES OF ORGANIC COMPOUNDS AT THE MELTING-
POINT.
Open Chain Compounds.
The hydrocarbons: n paraffins, olefins, and acetylenes.
The atomic volumes of carbon and hydrogen.
The highly constitutive nature of the melting-point probably
makes it generally unsuitable as a condition of comparison of
the molecular volumes of compounds. Nevertheless, in certain
cases, where the effect of constitution is constant, some remark¬
able regularities have been observed, and since the results in
general agree with those under other conditions, we may con¬
sider them to be significant.
In 1907 the author 1 dealt with a long series of normal
paraffins C n H 2 „ +2 , and showed that the volumes of the compounds
in the liquid state, at the melting point, form an approximately
arithmetical series, similar to that of their valencies. Thus if
V m represent the molecular volume of a member of the series
C*H 2n+2 , under the conditions, and W the sum of the valencies,
—C = 4, H= I—then the ratio V^/W is constant, and equal to
2*970 approximately. This is one-sixth of the difference for
CH 2 , viz. 17*82.
The data are those of Krafft, 2 who also showed that the
difference for CH 2 is constant, and equal to 17*80.
Digitized by
Google
2
LIQUID CHEMICAL COMPOUNDS,
TABLE I. —Thb Volumes of the Normal Paraffins.
Hydrocarbons .
W.
M. Vol.
= v m .
A for CH a .
V w /W = S.
W x 2*970
= M. Vol.
Undecane . . C 11 H S4
Dodecane . . C 19 H %
Iridecane . .
Tetradecane . . C 14 HjQ
Pentadecane . . C 15 H 89
Hexadecane . . C K B M
Heptadecane . . C 17 H 36
Octadecane . .
Nonadecane . . € 3S H tfi§
Eicosane . . C^H 48
Heneicosane . . C S1 H 44
Docosane . . Gjtl m
Tricosane . .
Tetracosane . . C 24 H M
Heptacosane . .
Hentriacontane . CgjH^
Dotriacontane .
Pentriacontane . CggH 7a
Mean Values
68
74
80
86
92
98
104
no
116
122
128
134
140
146
164
188
194
212
201*4
219*9
237*3
255*4
273-2
291*2
3090
326*9
344*7
362*5
380*3
398-3
416-2
4341
487-4
558-4
5762
629-5
• •{
18*5
17*4
18*1
17*8
18*0
17*9
17*9
17-8
17-8
17-8
18*0
17*9
17*9
3 x 17*8
4 x 1775
17*8
3 X 17-8
17-83
= 6 x 2*971
2*962
2*971
2*966
2-970
2*970
2 - 97 1
2*971
2*972
2*971
2*971
2*971
2*972
2*971
2*973
2*972
2*970
2*970
2*969
2*970
20196
219*78
237*6
255*42
273*24
291*06
308*88
32670
344*52
362-34
380* 16
398-oo
415-8
433-62
487*08
558-36
576-18
629*64
The above simple relation may be interpreted as follows:—
(a) The volumes of the atoms, carbon and hydrogen are
respectively the same in the individual compounds, but if only a
slight change occurs from compound to compound, these volumes
may be demonstrated by a comparison of a series of volumes.
( [b ) The volumes of carbon and hydrogen are
[H] = 2 970 and [C] = 4 x 2 970 = 1188,
and the relation (1:4) between these, is similar to that existing
between the valencies of carbon and hydrogen. The above
numbers are of course only average ones, but there is evidence
of only very small apparent deviations therefrom.
Also [CHJ = ii*88 + 5*94 = 17*82.
Direct Calculation of the Atomic Volumes of Carbon and
Hydrogen .
2[H] = [C^H*,] + [C 16 H3J - [C S1 H W ]
= 273*2 + 291*2 - 558*4 = 6*o
2[H] = [CjgHjJ - 18 [CHJ = 326*9 - 18 x 17*83 = 5*96.
The volume of hydrogen is thus very nearly equal to 3 0.
Also [C] = [CHJ - 2[H] = 17*83 - 6*0 * 11*83.
Digitized by v^ooQle
ORGANIC COMPOUNDS AT THE MELTING-POINT.
3
The relation between these values is
J§5 = = 4*0 (approx.).
[H] 30
These data give us the means of calculating the volumes of
the hydrocarbon elements in compounds at the melting-point.
[C„H 2n + J = »[C] + (2» + 2)[H]
= (6 n + 2)[H] = (6 n + 2) x 2*970.
Such values are found in the last column of the previous
table.
Krafft studied the molecular volumes of the above paraffins,
not only at the melting-points, but also at a number of tempera¬
tures up to ioo°C. This enables us to compare the volumes of
the paraffins at any temperature within these limits by making
use of interpolation formulae 3 of the type—
di = ff-r 1 ! - at ± fit 2 }, where t = T - T 1 .
The results are recorded in the following table:—
TABLE II. —Values of Vm/W illustrating the Additive Rule.
M.P.+ n x 10
CijHjh
CA
^14^80
^16^32
^16^34
CnH*
^18^38
M.P.
.
2*971
2'966
2-970
2*970
2*971
2*971
2-972
tt
+ IO°
2*997
2*993
2*994
2*997
3*000
2-998
2*999
tt
+ 20°
3025
3*020
3*023
3-024
3*027
3*025
3*026
tt
+ 30°
3*053
3*049
3*051
3*052
3*054
3*053
3*053
tt
+ 40°
3-081
3*078
3*081
3*080
3*082
3*081
3*082
tt
+ 50°
3*iii
3-107
3*109
3-iog
3* 111
3*iio
3* io 9
tt
+ 6o°
3*i4i
3*137
3*139
3*140
3*140
3*138
3*137
tt
+ 7°°
3172
3*i68
3*170
3*169
3-170
3-166
3*166
tt
+ 8o°
3-206
3200
3-201
3*200
3201
—
—
a
+ 90°
3*240
3*232
3*234
—
—
~
~
The above table shows that, at equal intervals of temperature
from the melting-points, and in the liquid state, the volumes of
the compounds are related to each other in a similar way to that
at the melting-points.
The effect of homology, or the effect of want of strict simi¬
larity between the volumes of the atoms in the different terms of
the series, is here at a minimum, and is in contrast to what we
find under other conditions, such as the boiling-point. The
nearly accurate manifestation of the additive rule extends to the
neighbourhood of ioo°, but under the latter condition there are
signs of a slight relative expansion among the more complex
members of the series.
1 *
Digitized by v^ooQle
4
LIQUID CHEMICAL COMPOUNDS.
i. The Influence of the Alternating Factor.
If the differences for CH 2 , as also the values V w /W, be ex¬
amined (see Table I), it will be found that they alternate in a
remarkable way among the earlier members of the series. This
is also found to be true of the melting-points, as was first pointed
out by Baeyer 4 in the case of the fatty acids and other series.
This is, however, a feature hitherto unrecognized in molecular
volumes, and it is evident that the peculiarity which causes the
melting-points of successive terms to alternate, also similarly
affects the molecular volumes. The data for compounds
simpler than undecane have been calculated by extrapolation
from Krafft’s data (loc. cit.), with the result that the effect has
been traced to simpler compounds, where it is more marked.
The melting-points of nonane C 9 H 20 and decane C 10 H 22 are
- 51 0 C. and - 2 5 0 C. respectively.
TABLE III.— The Effect of the Alternating Factor on the Molecular
Volumes of the Liquid Normal Hydrocarbons . 5
Vm
A between
A between
Substance.
W.
at
A.
Odd
Even
V w /W.
£
Tf
>
m.p.
Members.
Members.
C.H*
56
165*67
18*9
2*957
—
^10 H 22
62
184*54
169
35*7
—
2*976
ChHjh
68
201*40
18*5
35*4
2*962
—
CuHjj
74
219*90
17*4
35*9
—
2*972
Cia^aa
80
237*30
18*1
35*5
2*966
—
C14H30
C 15 H32
86
255*40
17*8
35*9
35*8
—
2*970
92
273*20
18*0
)
2*970
^16^84
98
291*20
17*8
35*8
—
2*971
104
309*00
2*971
It is apparent from the above table, that a difference in
volume exists between the odd and even members of the hydro¬
carbon series below hexadecane C 18 H 84 .
The odd members of the series have relatively small volumes,
but this depression gradually disappears as we ascend the series,
Digitized by
Google
ORGANIC COMPOUNDS AT THE MELTING-POINT 5
for in the table, the values of V w /W gradually increase to a
constant value—2*971.
The even members of the series have relatively large volumes,
and this abnormal expansion gradually diminishes as the series
is ascended until at about hexadecane C 16 H 34 the value of V w /W
is constant, or nearly so.
It follows that the values of V w /W, which respectively refer
to the odd and even terms of the series, can be arranged along
two curves, which respectively represent values which are either
depressed below, or raised above those which would belong to a
mean curve, or one for which V m /W is consistently 2*971. The
two curves are practically coincident with each other, and with
the mean condition, after C 16 H 34 . The alternating factor thus
exerts an opposite effect on an odd or even member, of the series,
these respectively possessing an odd and even number of carbon
atoms in the chain.
The influence of the alternating factor upon the volume fol¬
lows as a direct consequence of its effect upon the melting-point.
We have no reason to suppose that this alternating factor, which
may be considered to be due to a specific property of the mole¬
cules, specially affects the liquid volumes, but is only rendered
evident by differences between the melting-points. Since the
volumes are a function of temperature, they would naturally also
be affected. The result is, that the melting-points of the odd
numbers of the series are depressed and the volumes diminished
below a mean condition, whilst the melting-points of the even
members are elevated, and the volumes augmented above this
mean condition. The phenomenon just noted is doubtless
connected with the question of residual affinity. The members
of the even series evidently require a higher temperature for the
odd series in order to cause those changes which make possible
a relative motion of the molecules. Beyond this we know very
little. Biach 6 has endeavoured to account for the alternating
effects apparent in the melting-points of other series by means of
residual affinity. If we suppose that every pair of CH 2 groups
in the chains partially compensate each other, we see that there
would be an excess of residual affinity when the number is odd,
and a defect when even. This would cause a relative rise in
boiling point and diminution in volume in the case of the first,
and vice versd.
Digitized by v^ooQle
j6
LIQUID CHEMICAL COMPOUNDS.
The Effect of Unsaturation on the Molecular Volume at
the Melting-Point . 7
The detection of such a constitutive feature as the influence
of the alternating factor, suggests the possibility of unsaturation
under similar conditions, influencing molecular volumes. For¬
tunately the data are at hand to put this question to the test.
These refer to the olefins C n H 2 J = | and the acetylenes C„H 2n _ 2 |=|,
and are due to Krafft. 8 It will be convenient to draw our
conclusions simply from the differences, because the value of CH 2
is the same in the unsaturated series as in the normal paraffins.
This shows that the saturated atoms possess similar volumes in
saturated and unsaturated compounds, and that thus the effect
of unsaturation is only local.
TABLE IV.— The Value of the Effect for Unsaturation.
Substance.
V w at
A.
Substance.
Vm at
A.
m.p.
m.p.
Dodecane C lS H M
2199
87
Hexadecane CjgH^
291*4
8-5
Dodecylene | = ]
211*2
Hexadecylene C lft H M | = |
282*9
6*2
6*8
Dodecylidene |=|
205-0
H exadecylidene CjgHgo |= |
2761
Tetradecane C^Hg,,
2 55‘4
8-5
Octadecane C 18 Hgg
326*9
8-3
Tetradecylene C 14 H S8 [ = ]
2469
Octadecylene | = ]
3186
6-8
6-4
Tetradecylidene C 14 H 26 1=]
240-5
Octadecylidene J=J
311*8
1
It is apparent that unsaturation exerts a special effect on the
molecular volumes under the above conditions. This is doubt¬
less to be expected, since the molecules at the melting-point
acquire characters of fixity which precludes relative motion, and
because the molecular centres are less than at, say, the boiling-
point.
If we compare the volumes of the corresponding members of
any two series, we shall find that the differences are the same
throughout, and that those which exist between any two of the
terms in the three series are respectively some multiple of 17*8.
These facts indicate (a) that the effect of single and double un¬
saturation is constant, and ( b ) that the atomic volumes of carbon
and hydrogen are the same in the three series.
The effects due to unsaturation are calculated thus:—
Digitized by v^ooQle
ORGANIC COMPOUNDS AT THE MELTING-POINT.
7
-2[H] + M
= [C n H* n | = |- CnH^+J = - 8-5
But 2[H] = 2 x 2*97o = 5*94.
Therefore | = ] = - 8*5 + 5*94 = - 2*56.
Also - 4 [H] + [=]
= [C«H an _ J [=| — C n H an + 2 ] — = 15*1
But 4[H] = ii*9
.*. |=| = - 15*1 + n-9 = - 3*2.
The depression caused by an olefin linkage | = | is equal to
- 2*56, and that for an acetylene linkage |=| is - 3 2.
These results show that a special effect due to unsaturation is
observable at the melting-point, and this is in agreement with
the conclusions drawn from the observation of other physical
properties. One difference, however, is that, whereas, in optical
refractivity and magnetic rotatory power, an increase in the
magnitudes of the constants is observable, in molecular volumes
on the other hand a decrease is noticed. These differences are
what we should expect. It is to be supposed that the nearer
that absolute zero is approached, constitutive effects like these
become more marked. There is also the possibility that at the
boiling-point, when the distance between the molecular centres is
greater than at the melting-point, the effect for unsaturation may
disappear altogether. The melting point is a highly constitutive
property, and these temperatures are not related to each other
in the same way as the boiling points of compounds. For this
reason, the volumes referred to the melting-point, are not as a
rule comparable with those referred to the boiling-point. The
constitutive factors vary least from compound to compound,
among the normal paraffins.
The Nitriles.
In addition to the hydrocarbons, and particularly the un¬
saturated ones, Kraflft 9 has also studied the alkyl cyanides or
nitriles at the melting-point. They are unsaturated in a similar
sense to the acetylenes, as a comparison of their formulae will
show;
R - C = CH R - C = N
Acetylenes. Nitriles.
The volume of combined nitrogen is—
[=N] = 3 [H].
Digitized by v^ooQle
8
LIQUID CHEMICAL COMPOUNDS,
On this assumption, we shall see from the following table,
that there is an outstanding difference for unsaturation, on com¬
paring the observed and the calculated values.
TABLE V. —The Volumes op the Nitriles.
R - C = N.
Substance.
W(R).
V m .
A for CH a
Vol. R.
- C = N.
Lauronitrile C 11 H as C = N
67
216*8
199*0
17-8
Myristonitrile C = N
79
252*3
35*5
2346
17*7
Palmitonitrile C 16 H S1 C = N
9 i
288*2
35’9
270*3
17-9
Stearonitrile C = N
103
324*0
35-8
305 9
18-1
Mean Values .
•
u
35’7
x 17-85
17-87
The group - C = N evidently equals that of CH 2 in volume.
On the assumption that
[C] = 4 [H]and[N]==3[H]
- C EE N = 7[H] = 20*8 (calc.)
A for |=[ = 17*87 - 20*8 — - 2*93.
Oxygen Compounds: Ketones and Fatty Acids.
(Krafift, loc. cit.)
Hydroxylic Oxygen (OH).
The melting-point is a physical property very sensitive to
constitutive influences, and it is necessary to obtain the substance
quite pure in order to show a melting sharply at one temperature.
Slight traces of impurity are a serious impediment to an accurate
determination. Some impurities are particularly difficult to get
rid of, such as small quantities of substances belonging to the
same series as that of the compound which is being examined.
It follows that an accurate study of molecular volumes is de¬
pendent on an accurate estimation of the melting-points of the
substances. How far the discrepancies, observed in a study of
the molecular volumes, are due to such inaccuracies, it is difficult
to say, but some may be due, at least in part, to this cause.
In drawing up a table of the alcohols we assume, what is very
probable, that in each compound the atom of hydroxylic oxygen
is equal in volume to twice that for a single hydrogen atom in
the same compound, i.e.
[O'] = 2[H].
Digitized by v^ooQle
ORGANIC COMPOUNDS AT THE MELTING-POINT
9
TABLE VI.— The Normal Alcohols.
Le Bas (loc. cit.).
Compound.
w.
v*.
Vm/W.
W x S
S a=S 2*970.
Nonyl Carbinol
c 10 h 21 oh
64
188*3
2*943
190*08
Undecyl „
CjjHjjOH
76
223-9
2-946
225*72
Tridecyl ,,
C^H^OH
88
259-8
2-953
26l*36
Pentadecyl „
Cj.HjjOH
100
296*0
2*960
297*00
Heptadecyl „
c 18 h„oh
112
332*3
2*970
332 - 64
An examination of the numbers shows that it is only at
about heptadecyl carbinol that the value of V^/W approximates
to the normal value of 2 970.
There is a variation in the volumes as we ascend the
series, which variation may be considered to be similar to those
which we find at the boiling-point and due to the addition of
the homologous increment CH 2 . The significance of this prob¬
ably is that, for the particular series in question, the compounds
are not quite comparable at the melting-point. This want of
correspondence between the two series—the melting-points and
the molecular volumes—involves a difference between the volumes
of the carbon and hydrogen atoms in the compounds from
their usual values, derived from the normal paraffin series.
The temperatures at which the volumes of these atoms are
normal, are evidently not the melting-points. The value of
hydrogen which has been noted, viz. : 2 970, is, however, likely
to occur at the melting-point at some point of the series which
is being examined. Among the alcohols, this occurs at hepta¬
decyl carbinol C 18 H 87 OH.
Ketonic Oxygen (C : O).
The melting-points of the compounds of this series show signs
of alternation, and this fact may possibly influence the molecular
volumes somewhat.
The data, except for the more complex compounds, are of *
miscellaneous origin. The latter are due to Krafft (loc. cit.).
It is remarkable that, whilst the simpler compounds show
values of V,»/W, which fall below the normal value 2 970, the
Digitized by v^ooQle
IO
LIQUID CHEMICAL COMPOUNDS.
most complex show values which are in excess of this. Com¬
pounds of the complexity of (C 6 H 13 ) 2 C : O, in their results, are
similar to those of the paraffin series, in that the values of V m /W
are = 2*970 exactly.
[O"] = 3[H] = 3 x 2-970 = 8-91.
TABLE VII. —The Volumes of the Ketones.
w.
Compound.
m.p.
v m .
A.
W x 2*970.
A.
63
C 8 H„ . CO . CH,
3*5
186-2
_
187-1
+ 0-9
69
C 9 H 19 . CO . ch 8
15-0
204-4
—
204-9
+ 0-5
69
C 6 H u . CO . C 6 H n
14*6
204-3
—
204-9
+ o-6
81
CgHu • CO . C 6 H 13
28-0
240-6
240-6
_
35*6
93
C 7 H, s . CO . C,H, S
40-0
276*2
2 x 17-8
276-2
—
35-8
105
c 8 h 17 . CO . C 8 H 17
49*o
312-0
311*8
- 0*2
141
Laurone
69-0
420-6
_
418-8
-i-8
^11^28 • CO . CnHjtf
165
Myristone .
71*1
C 19 H„.CO.C lS H„
76-3
491-7
4 x 17-8
490*0
-17
189
Palmitone .
71-0
CibH 81 . CO . C 15 H 31
8a-8
562-7
56 i *3
-i*4
Stearone .
71-6
213
CnHgj . CO . Cj.Hgg
88-4
634-3
632-6
-i*7
Mean value
• • -(
71-2
\
X
»
M
II
4
It is probable, from the evidence of the above numbers, that
the series of melting-points and the series of temperatures for
which the volumes of similar atoms are equal, are not quite the
same. We notice that the differences between the calculated
and observed molecular volumes for laurone, etc., are the same,
viz. I *70.
The following results show that
[O'T = 3[H]
in the ketones, and that the apparent deviations therefrom, are
due to disturbances of the total volumes of the compounds, for
some reason similar to that already stated.
Vm O"
[C 16 H3oO] + 2[H] = 282-17
897
[CuHgJ = 273-2
and [C^H^O] + 2[H] - 317-97
8-97
[CjtHjjJ = 309*0
Digitized by v^ooQle
ORGANIC COMPOUNDS AT THE MELTING-POINT. n
This is sufficiently near to 8*91 = 3 x 2*970 = 3 H to make
it probable that the volume of O' in the ketones is in reality equal
to 3 H.
Carboxylic Oxygen C : O (OH).
The fatty acids C n H 2rt 0 2 are subject to the alternating factor,
but sufficient data do not exist for a study of this subject in
relation to them.
Krafft, whose excellent experimental work has already been
referred to, has extended his observations to a few mono-
carboxylic fatty acids, which are connected with the compounds
already studied. They correspond to those ketones which were
shown to be normal in their atomic volumes.
TABLE VIII.— Volumes of the Fatty Acids.
Compound.
W + 1.
V m at m.p.
(Krafft).
A.
V
W + 1.
Lauristic.
Myristic .
. C 12 H M O a
77
228*5
35*9
2*970
89
264-4
35*8
2*970
Palmitic .
IOI
300*2
2*972
357
Stearic .
113
335*9
2*972
Mean Value
■ • (
35*8
1 =
2 x 17*9
The above results indicate that amongst the carboxylic acids
the relations [O"] = 3[H] and [O'] = 2H hold.
Volume of C 0 2 .
This may also be deduced in the way shown in the following
table;—
TABLE IX.
Fatty Acid.
v m .
Paraffin.
Vm.
CO a .
Lauristic
• C^O,
228*5
C„H m
201*4
27*1
Myristic .
264*4
CuH*
237*3
27*1
Palmitic .
• c 16 h w o 2
300*2
CbH*
273*2
27*0
Stearic .
• C w H le O,
335*9
c, 7 h*
309*0
26*9
Mean Value
.
27*0
Digitized by v^ooQle
12
LIQUID CHEMICAL COMPOUNDS.
g[H] = 9 x 2*970 = 26*73.
It follows that [C 0 2 ] = 9[H], and since [C] = 4[H], then
2[0] = 5[H] =15 x 2*970 = 14*85.
Volume of 0 2 .
This is also shown to be the case by the following calcula¬
tions—
Vm. O..
[Ci 4 H m OJ + 2[H] 270*37
14*97
[C^Hig] 255*4
[CigH^OJ + 2[H] 306*17
14*97
[CieHjJ 291*2
The value obtained by difference, is similar to the one calcu¬
lated. It is approximately equal to 5[H]. Remembering that
[O"] = 3[H], in the ketones, and that [O'] =■ 2[H] in the alco¬
hols, we account for the value of O in the group - C^qjj
as follows:—
[O"] + [O'] = 3 [H] + 2[H] = 8*91 + 5-95 - 14*85.
The carboxyl oxygen in the fatty acids thus possesses a similar
volume to that of carbonyl oxygen in the ketones , and the remaining
oxygen is hydroxylic as in the alcohols.
The latter may be calculated independently as follows:—
Vm. Vm.
[CjjH^OJ (fatty acid) 228*5
[CjsHaeO] (ketone) 2 4°’ 6 \ 2 22*8
less [CHJ - 17*8J 222 *
A for [O'] = 5*7(2 x 2*95).
Whilst the results are what we may call normal for com¬
pounds of about the complexity indicated (C lc ), simpler com¬
pounds are evidently different. The reason has been indicated.
Note on Pelargonic Acid.
Besides those already noticed, we may remark that KrafIVs
value for pelargonic acid (C 18 H 17 . COOH) of melting-point 12-5
is 173*2, or —
Vm = 173*2 observed. 2*Va 175*2.
This is a number which is larger than the observed value by
175*2 - 173*2 = 2*o, and consequently the latter is depressed to
this extent below the normal. If, however, we compare the
volume of pelargonic acid with a compound (alcohol) of similar
complexity, we observe the following results :—
Digitized by v^ooQle
ORGANIC COMPOUNDS AT THE MELTING-POINT. 13
Alcohol.
Vm. Acid.
V w .
C, 0 H„O'
188-3 C 9 H 18 0 ' 0 "
173-2
plus CH 2
17-8
2SnVa
191*0
Vm
191*0 C 10 H„O"
follows that
[ 0 "] = 3 [H] since [O']
= 2[H]. Then
[C 10 H 22 O"] - [C 10 H 22 O'] = [O"] - [O']
= 191*0 - 188*3 = 27
which is a number but little different from the theoretical
difference.
CO"] - [ 0 ] - [H] - 2*97.
Conclusion .
We thus conclude that if at a certain point in a series
the volumes of the carbon and hydrogen atoms are similar to
their normal values, which have been deduced from a long list
of normal paraffins from C 1€ H 34 to C 35 H 72 , the values for simpler
compounds will fall short, and those for more complex compounds
will be in excess of these. This indicates that there is probably
a progressive increase in the atomic volumes of, say hydrogen, as
measured by the ratio V m /W from quite the beginning of a series.
This effect is, however, complicated by the alternating factor in
alternating series. We shall observe what is possibly a similar
result at the boiling-point, and it may be supposed to be due to a
want of strict similarity between consecutive members of a series
under these conditions. The mode of increase at the melting-point
is, however, not quite so clear. In conclusion, we may note that
it is probably true, that the molecular volumes of the various
compounds in different homologous series, are not quite com¬
parable at the melting-point, and since the melting-points are
not related to one another in the same way as the boiling-points
of the same series of compounds, we must expect differences in
the variations of the volumes of a homologous series at the two
points respectively.
As regards the atomic volumes in the carboxylic group, we
have found that—
[COOH] = [C] + [H] + [O'] + [O"]
= ii*88 + 2-97 + 5-94 + 8-91
= 29-7 (calculated)
[COOH] = 30*0 approx, (by difference).
Digitized by v^ooQle
14
LIQUID CHEMICAL COMPOUNDS.
Note .—In order to show how the observed volumes may
vary—for reasons unknown at present—Schiff, a very careful
and accurate worker, found a value of 333 for stearic acid
C^HggCOOH, whilst Krafft, on the other hand, found the volume
to be 335*9 for this compound, and in view of the fact that the
latter took special care to experiment with pure substances, we
must give to his value at least equal weight to that which might
be accorded to the other. Moreover, the larger number accords
with those obtained for the neighbouring homologous fatty acids
by Krafft This raises the question of why workers of equal
reliability should obtain different experimental results, which
differences, indeed, seem too large to be ascribed to the effect of
impurity, but this can only be settled by very careful and accur¬
ate experimental work on substances prepared in several different
ways.
Ring Compounds.
Schiff 10 determined the volumes of a number of aromatic
compounds which are possessed of ring structure, in order to
find out whether significant volume relations were to be met
with, similar to those which Krafft had pointed out for paraffin
derivatives at the melting-point. Schiff failed to find such re¬
lations, and indeed any significant relations whatsoever.
The reason no doubt is, that the melting-point is a highly
constitutive physical property, and if we are to consider the
volumes of substances in the liquid state at the melting-point,
we must expect large deviations from the values which might be
expected from a consideration of their composition alone.
If the compounds are in the liquid state at the melting-point,
and we suppose that the structure of liquids does not change in
the interval from the boiling to the melting-points, we must
suppose that they are subject to the law of coincident states.
Under such circumstances we might expect, that a condition of
comparison would be found at equal fractions of the boiling-
point. The melting-points are not equal fractions of the boiling-
point temperatures of the compounds, and thus they do not
represent a suitable condition of comparison, at least from the
same point of view.
The following investigation shows, that the volumes of com¬
pounds are comparable at approximately equal fractions of the
boiling-point temperatures:—
Digitized by v^ooQle
ORGANIC COMPOUNDS AT THE MELTING-POINT . 15
TABLE X.
Compound.
m.p.
or
0°.
b.p.
1 _ m *P*
b.p.
= A.
d m.p*
d b.p*
dm.p. _ j
db.p-
= B.
Ratio
B/A.
C.H.
6°
8o°
0*210
0*8940
08133
0*0980
0*470
c 4 h 4 s
Cio H 8
o°
V
0*236
1*0884
0-9874
0*1023
0434
79 °-2
217 0
0*281
0*9777
0*8674
0*1270
0*456
c 10 h i4
o°
200°
0*423
0*9419
07809
0*2062
0*487
C,4H 10
. ioo°*5
340 °
0*391
1*0630
0*9073
0*1720
0*440
c 5 h 5 n
o°
II 5 °
0*296
1*0033
0*8826
0*1367
0*462
CjH,N
C„H«. C a H 4
o°
234 °
0*462
1*1081
0*9211
0*2030
0*439
103°
277 0
0*316
1*0300
0*9018
0*1422
0*450
Mean. . .
0*455
Conclusion .—This table shows, that at equal fractions of the
boiling-points the molecular volumes of the compounds are
similar although not exactly equal fractions of the volumes at
the boiling-points. The nearly constant relation between A and
B, indicates that there is a relation between the changes in
volume of the molecules, and the changes in temperature, which,
so long as the liquid state is maintained, is approximately the
same for the ring compounds mentioned and others of similar
type. The condition necessary for the volumes at the melting-
points to show similar results relatively to each other, as at the
boiling-points, is that A should be the same for all, or that the
interval between the melting- and boiling-points should, in all
cases, be the same fractions of the boiling-point. This condition
is not realized, hence the volumes at the melting-points are
subject to all the large and preponderant constitutive influences
of the melting-points themselves.
The molecular volumes of compounds at the melting-point
cannot thus be utilized in the same way as those at the boiling-
point.
General Conclusions.
A number of striking relations have been made out from a
study of the open chain hydrocarbons, saturated and unsaturated,
and some of the derivatives of the normal paraffins. These in¬
vestigations, while intrinsically interesting, do not cover sufficient
ground to enable us to draw up a list of rules which might be
of use in the study of compounds of unknown structure. More¬
over the nature of the limitations of the additive rule are not
Digitized by v^ooQle
i6
LIQUID CHEMICAL COMPOUNDS .
clearly understood, even in open chain normal compounds.
Still less is this so in the case of compounds of more complicated
structure, such as ring compounds. This must be ascribed to
the highly constitutive nature of the melting point, a feature
which would be likely to make the elucidation of the constitutive
influences acting on molecular volumes at present a difficult
matter. The results obtained, however, show, that in some cir¬
cumstances, additive relations may be clearly revealed, and what
is important to remember, the relative volumes of the atoms
are similar to those at the boiling-point. The following table
contains a list of the chief results obtained.
TABLE XI.— Table of General Results.
40>4
Atom or Linking.
Value.
Atom or Linking.
Value.
C
n-88 4 [H]
- CH : CH 2 | = |
- 2*56
H
2*970
-c ; ch - |=|
- 3*20
O'
5*940 2[H]
-c : n - i=i
- 2-97
O"
8*910 3 [H]
N"'
8 910 3[H]
Digitized by v^ooQle
CHAPTER II.
A STUDY OF THE HYDROCARBONS AT THE BOILING-POINT.
The Open-Chain Hydrocarbons.
The Volume of Carbon and Hydrogen .
These atoms are of fundamental importance in a study of the
molecular volumes of organic compounds, and as such demand
careful consideration. This is especially necessary in view of the
fact that Kopp’s numbers 11 are unreliable.
We may follow his line of reasoning, by a study of the
compounds butane C 4 H 10 and benzene C a H 6 , according to his
method. Thus:—
[C 4 H 10 J = [C 6 HJ = 96-0 . . . (i).
It was also shown that
[CHJ = 22*0 . . . (2).
From (i) we conclude that [C] = 2[H], and applying this to
(2), we find that
[C] = iro and [H] = 5*5.
The compounds butane C 4 H 10 and benzene C a H a , and also
those actually studied by Kopp, belong to two entirely different
classes of chemical compounds, and are possessed of quite
different structures. Butane is an open-chain compound and
benzene possesses ring structure. It is thus not necessarily true
that similar atoms, are identical in volume, in the two classes
of compounds mentioned. This was the assumption Kopp made,
and, as we shall see, it was not justifiable.
The most direct way of showing this would be to ascertain
the volume of dipropargyl C a H a , a compound which has a struc¬
ture similar to that of butane and a composition similar to that
of benzene. Its molecular volume has not been directly deter¬
mined, but the probable value can be obtained quite easily,
from a study of the available data for the unsaturated hexane
derivatives.
17 2
Digitized by v^ooQle
i8
LIQUID CHEMICAL COMPOUNDS.
TABLE XII.
Compound C 6 .
Vm.
»H,.
Hexane
CH, . CH, . CH, . CH, . CH, . CH,
139*9
7*5
Hexylene
CH, . CH, . CH, . CH, . CH : CH,
132*4
2 x 6*9
Diallyl
CH, : CH . CH, . CH,. CH : CH,
126*1
4 x 7*4
Dipropargyl CH : C . CH, . CH, . C j CH
[n°'3]
The first three compounds have been directly investigated;
the fourth value is found by extrapolation, assuming the validity
of the summation law.
The volume of dipropargyl:—
CH : 0 . CH, . CH, . C : CH is 110*3.
The volume of benzene
CH
^ \
CH CH
is 96*0
:h ch
\ /
CH
- 14*3
We conclude, that the volumes of similar atoms, are not the
same in the two compounds, nor in the two classes of com¬
pounds, and consequently Kopp’s assumption is unjustified. It
is seen that on passing from open to closed-chain structure, there
is a large contraction of - 14*3 which must be accounted for.
The next step, is to show what the actual values of carbon
and hydrogen in paraffin and benzene derivatives are.
If we take the same two compounds—butane and benzene—
into consideration, and for the moment, suppose that the volume
represented by [C*H c ], is that of benzene, but with its atoms
similar in volume to those of dipropargyl, then—
[C 6 HJ - 4[H] = 96*0.
But [C 4 H 10 ] =s 96*0.
It follows, that [C 4 H 10 ] = [C e HJ - 4 [H]
and thus [C] = 4 [H] . . . . . (1)
But [CHJ = 22*2.(2)
Therefore [C] = i 4 *7 and [H] = 37.
It is interesting to note that the relation [C] = 4[H] is the
same as that which was also found to be true at the melting-
point
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT 19
The values [C] = 147 and [//] =37 are the actual atomic
volumes of combined carbon and hydrogen at the boiling-point
probably in all compounds with open-chain structure . At any
rate we must consider them to be the normal values .
The conclusions of contemporary writers were not always
the same as those of Kopp, for Lothar Meyer advocated a value
of 3*0 for hydrogen, and Loschmidt came very near to anticipat¬
ing the true values by the numbers [C] = 14*0 and [H] = 3*5.
The benzene difficulty was got over by assuming that the
volume of each of three carbon atoms was equal to 11 *o, and that
of each of the other three 14*0. The volumes of the hydrogen
atoms were given as 3*5.
Reverting to the numbers which are now proposed for com¬
bined carbon and hydrogen, we see that they stand in the
relation of 4 :1.
It can be shown that on passing from butane or dipropargyl
to benzene the relation already indicated is preserved unimpaired.
Thus, at the critical point, the volumes of carbon and hydro¬
gen are
[C] = 387 and [H] = 9*67
among the paraffins—values which are as 4: 1.
For ring compounds, a different result is obtained.
Benzene [C 6 HJ = 256*3
Hexamethylene [C 6 H 12 ] = 306*7
A for 6[H] 50*4 = 6 x 8*40
6[C] = 256*3 - 50*4 = 205*9
and [C] = 34*3 or 4 x 8*57,
[H] » 8*40.
These values are also approximately in the relation of 4: 1.
By assuming this relation to be true, and dividing the critical
volumes of benzene and hexamethylene by the respective valency
numbers, we obtain nearly constant values for V^/W, which values
represent the volume of hydrogen.
v c w. v c /w
m ml
Benzene C 6 H 6 256*3 30 8*54
Hexamethylene C 6 H 13 306*7 36 8*52
The 4: 1 rule may be shown to be true, not only at the boil¬
ing and critical points, but also at all equally reduced pressures.
2 *
Digitized by v^ooQle
20
LIQUID CHEMICAL COMPOUNDS,
We conclude from this, that on passing from the aliphatic to
the aromatic class of compounds, the volumes of the carbon and
hydrogen atoms undergo a contraction, but in such a way as to
preserve the characteristic numerical relation unimpaired The
total contraction for benzene and the other ring compounds is
equal to the sum of the contractions of all the atoms in the
nuclei.
TABLE XIII.— Direct Investigation op the Atomic Volumes op Carbon
and Hydrogen. See Le Bas (loc. cit.) and ref. 12
Substance.
Vw.
C 6 H u
117-8
C.H 14
139-9
2[H] = [C 5 H,J - 5[CHJ = 117-8 - 5 x 22-i = 7-3 (2 x 3-65)
[C] = 22-i - 7-3 = 14-8 (4 x 3-70).
The above results are thus in favour of the average atomic
volumes already given—[H] = 370 and [C] = 14/8, and the
evidence seems to be indisputable. The result of such evidence
is to completely invalidate the original numbers of Kopp.
Those which have just been obtained, are, however, also sup¬
ported by a comparison of the volumes of certain normal paraffins,
olefins, diolefins, etc.
TABLE XIV. —The Atomic Volume op Hydrogen.
Substance.
Vm,
»h 2 .
Vm.
Olefin.
C»H la
117-8
77
iio-i j
c 5 h 10
—
—
2 x 6*7
104*5
C,H 14
139*9
7*5
132*4
C,H la
—
2 x 6-9
126-1
C,H 10
c*h 7 ( 0 H)
81-5
7*3
74*2
CjH s (OH)
C,H,C 1
917
7*3
84-4
C s H 6 C 1
N(C3H,) 3
222*4
3 x 7-2
200*7
N(C s H 5 ) 3
Mean value of [H] a = 7*15
A similar result is obtained from a consideration of the
volumes of ethyl benzene, and its unsaturated derivatives.
Compound.
Vm.
C,H t .
CHj.
CH S
139*3
c,h b .
, CH 2
CHj
131*5
c,h 5 .
C :
CH
126*2
A for nH a .
7-8
2 x 6-6
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT
21
The above compounds are for the most part unsaturated,
but similar results can be obtained from a consideration of
saturated ones. Thus—
H
H
COOCH3
COOCH3
= H«
COOCHg
Aooch*
Two molecules of methyl formate. Dimethyl oxalate.
2[H] = a[H . COOCHJ - [(COOCH,)J = 125-0 - 117-4 = 7’6.
It follows from the above calculations that the average
volume of combined hydrogen in organic compounds is—
[H] = 370.
TABLE XV.— The Atomic Volume of Carbon.
Substance.
Vm.
A for Carbon.
Vm.
Olefin.
C 4 H 10
96*0
14*1
110*1
C 6 H io
117*8
14*6
132-4
C,H U
c,h 14
139*9
14*9
i 54'8
c,h ]4
C,H 16
162*6
15*0
177-6
c 8 h 16
Mean value of carbon [C] = i4*6
The average volume of carbon, taking all the data into con
sideration, is—
[C] - 14*8.
The volumes of combined carbon and hydrogen given above,
are average ones only, and, as such, are of no special significance.
They are, however, useful for purposes of calculation.
If we except liquid hydrogen itself, and perhaps methane, the
combined volume varies within the comparatively narrow limits
of 3-4, even including compounds of widely different structure.
Note on the Effect of Unsaturation at the Boiling-point. —A note
is required on the result which has been obtained for unsatura¬
tion. It is seen that the values for corresponding members of
the n paraffin, olefin and acetylene series are simply dependent
on their composition. There is thus no special effect due to
unsaturation at the boiling point . The evidence for this conclu¬
sion is so strong that it is scarcely open to question.
Digitized by v^ooQle
22
LIQUID CHEMICAL COMPOUNDS.
Very different results have previously been obtained by
different observers. Buff 13 , in 1865, sought to show that carbon
possesses greater volume in unsaturated than in saturated com¬
pounds. This was apparently confirmed by Schiff, 14 who calcu¬
lated that the increase for single unsaturation, was 4 0 units,
approximately. Thus—
v w .
Amylene C 5 H 10 no*o
Pentane C 5 H 12 117*2
A = 7*2
Then if H 2 = 2 x 2*55
1 = 1 = n*o - 7*2 = + 3*8.
Lossen 16 thought that this increase was due to the hydrogen
rather than the carbon.
These results are evidently due to the particular atomic
volumes of hydrogen and carbon chosen.
If we grant the validity of the new numbers, the obser¬
vation, that unsaturation is responsible for no special effect, follows
as a matter of course.
The Influence of Complexity on Molecular Volumes , especially that
connected with the Influence of the Homologous Increment .
It has been repeatedly stated that the additive rule is not
quite followed in molecular volumes. This was first shown by
Gartenmeister, 16 Dobriner, 17 Schiff, 18 Pinette, 19 Zander, 20 and
others, from the study of the compounds of various homologous
series, that is, compounds which differ in constitution by the
group methylene (CH 2 ). The series in question are those of the
ethereal salts of the fatty acids, the ethers, the alkyl iodides, etc.
More recently, Young 21 has shown, that, under a variety of
conditions, the normal paraffin series of compounds, always differ
from term to term when the group CH 2 is added. Thus, under
no circumstances, is the additive rule exactly realized.
The differences in volume between two successive terms of a
series, have, however, been shown to give unreliable information
regarding the extent to which the additive rule is departed from.
The accompanying table shows this.
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT
23
TABLE XVI. —The Volumes of the n Paraffins at the Boiling and
Critical Points.
Substance.
W.
V m
at the
boiling-
point.
A.
v m /w.
v*.
m
A.
v e
m
W*
n Butane C 4 H 10
26
96*0
3-693
251*0
9-654
21*8
58*3
n Pentane C 6 H 12
32
117*8
3-681
309-3
9-666
22*13
56*8
n Hexane C 6 H 14
38
I 39*93
3-682
366*1
9-634
22*63
6o*2
n Heptane C 7 H 16
44
162*56
3*693
426*3
9-689
If we take into consideration the fact that in the aldehydes
the volume of oxygen is equal to twice that of hydrogen, and
that, so far as a comparison can be made, the members of the two
series possess almost identical volumes, we do not perhaps com-
mit any very serious error, if we suppose that the volumes of the
simplest paraffins, are similar to those of the aldehydes, which
have been investigated.
TABLE XVII. —The Volumes of the Normal Paraffin Series.”
Substance.
w.
V w .
Vm/W.
ch 4
C.H,
8
38-5
4*812
14
56-7
4*050
c,h 8
20
74-6
3*730
c 4 h m
26
96*0
3*693
C 5 H la
32
117*8
3*681
C.H„
38
139*93
3*682
CtH 16
44
162*56
3*695
c„h 18
50
186*26
3*725
The curve shown on next page is similar to those obtained for
many series and indicates the variations in the volume of combined
hydrogen in the series. It consists of three parts:—
(а) A descending arm : This shows that the volumes of
hydrogen diminish rapidly from compound to compound.
( б ) A minimum: The volumes of hydrogen are a minimum at
this point. For a limited range of complexity the atomic
volumes do not differ much. It is to this circumstance that the
approximate additive relations which have been noted, are due.
(c) An ascending arm: The values of V/W increase, possibly
in a rectilinear way.
Digitized by v^ooQle
24
LIQUID CHEMICAL COMPOUNDS .
The influences which contribute to the above result, are not
quite clear, but we are probably correct in supposing that several
are operative, The following explanation seems to be at least
plausible. We suppose, that, at any rate, the following two
influences are probably important: (a) the influence of the element
carbon with its enormous power of self affinity; ( b ) and the effect
of the volatile hydrogen with its large molecular volume. These
Fig. 1.
characteristics are not lost in combination. It follows, that the
volume of methane, might be expected to be relatively large, owing
to the large proportion of hydrogen atoms present in the mole¬
cule, compared with the number of carbon atoms. As methylene
groups (CH 2 ) are added, the proportion of carbon increases, and
thus there is a very considerable diminution in the values of
V m /W. The reason for the ultimate increase in the value V m [W
is not quite so clear. This possibly is owing to a gradually
changing relation of the volumes to the boiling-points, which also
involves the magnitude of the internal pressures due to the
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT.
25
intramolecular forces. Thus in general terms, we suppose that the
intramolecular and consequently the intermolecular forces are
different among the different compounds of a homologous series,
so that the molecules are freed at temperatures which do not
preserve the necessary volume relationships which make the
compounds completely homologous. The extent by which they
differ indicates the constitutive differences between them. Young
attributes these variations to constitutive causes.
The Branching of the Hydrocarbon Chain .
This is the only constitutive influence likely to affect the
volumes of the open-chain hydrocarbons, since unsaturation does
not cause any special effect. The data are, however, very meagre,
so that the conclusions have not that proved generality which is
desirable.
The iso structure seems to cause a definite lowering of the
boiling-point (8-9 degrees), and a definite contraction (0*50).
TABLE XVIII. —Contraction for the Iso Group.
Normal Compound.
Vm.
Iso Group.
V w .
Iso Compound.
n Pentane C 6 H 12
117*8
- 0*40
117*4
(CH s ) a . CH . CH a . CH,
n Hexane C 8 H l4
139*9
-0*50
139*4
(CH 3 ) 2 . CH . CH 2 . CH 2 . CH,
n Heptane C 7 H 16
162*9
-o*6o
162*3
(CH s ) 2 . CH . CH,. CH a . CH 2 . CH 3
n Octane C 8 H J8
186*6
- 0*90 x 2
185*2
(CH 3 ) 2 . CH . (CH a ) 2 . CH, . (CH 3 ) 2
Valerylene C 5 H 8
104*2
-0*30
103*9
Isoprene
A =
- 0*53
The contractions due to the presence of the iso group, seem
to increase with complexity, and the average value is about
(Iso group) 0*50.
The observed contraction, which is characteristic of com¬
pounds with the iso group, is no doubt associated with a greater
concentration of matter at the point at which this group is
situated, than at other points in the molecules. This is shown
by the following scheme—
Digitized by v^ooQle
26
LIQUID CHEMICAL COMPOUNDS, m
(i)
C
C
The iso group.
( 2 ) —C—C—C—C—C—
Normal chain structure.
The contractions which have just been indicated, illustrate
that great capacity for self-affinity which characterizes carbon,
and which is most effective when the atoms are most concentrated.
It is possible that the large contractions which distinguish ring
structure, may be due to this cause. This more compact arrange¬
ment, affects not only the carbon, but the hydrogen atoms also.
Naumann 23 in 1874 stated the following rule for the boiling-
points of the paraffins and other open-chain hydrocarbons: “ The
more nearly the grouping of the atoms in an isomeric paraffin
deviates from the rectilinear or chain type> and approaches the
spherical type y the lower is the boiling-point
A similar rule, moreover, may perhaps apply to the mole¬
cular volume, that is, so far as observation goes, and the rule
may have as great generality for volumes as for boiling-points,
provided we limit it to the various hydrocarbon series.
The Interaction between Methyl Groups .
The average value of the contraction for an iso group is
-o*5, and one of about twice the magnitude may be deduced
from the volume of di-isobutyl which contains two iso groups.
For di-isopropyl, where the iso groups are nearer to each other, a
much larger contraction is noticed.
Di-isobutyl.
c 8 h 18
Di-isopropyl. C 6 H 14
ch 3
CH,
CH, CH 3
d:H—CH 3 —CHa—CH CH—CH
ch 3 ch 3 ’ ch 3 Ah 3
A - 1*4 (2 x - 07).
A -3*1.
Vm 185*2.
136*8.
The magnitude of A in di-isobutyl is equal to the contraction
caused by two iso groups acting independently, but this double
value is augmented in di-isopropyl, owing to the closer approxi¬
mation, and consequently probable interaction of these groups.
The augmentation in question, is—
- (3*1 - 1*4) - - 17.
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT.
27
The real cause of the above contraction is considered to be
due to the interaction of CH 3 groups attached to different carbon
atoms. This is evident from the diminished boiling points.
If we consider a compound in which an iso group and an
ethenoid linkage are close to one another, we shall notice a
disturbance in the total value of the constitutive effect.
v m .
n Amylene
C 3 H 7 - CH = CH a
110-5*
* [This value has been calculated by subtracting 7*3 from the molecular volume
of pentane Vm = 117*8.]
A - 0*40 This is about equal to
one iso group.
iio*i (Sch.)
A - 2*40
108*1 (Th.)
CH 8
Reasons are given later on for supposing that the above con¬
traction is due to the interaction of two CH 3 groups. This
necessitates a modification of the above formula, thus—
CH 8 — c -CH 8
11
CH—CH 8
Other compounds of this type are studied later on. The rule is
that when two methyl groups (or four) are attached to neighbouring
carbon atoms there are contractions due probably to interaction
between these groups.
In the foregoing account, we have dealt entirely with the
influence of the iso group, but of course, a hydrocarbon chain
may be divided up in different ways.
a Amylene
CH 8
^H-
CH = CH,
0 Amylene
CH 8
i =
CH - CH,
The Influence of the Tertiary Grouping on Volume.
Doubtless the grouping (CH 3 ) 3 C - would show an increased
contraction, as compared with an iso grouping, but actual data
are lacking for the hydrocarbons. The volume of pinakolin
(CH 3 ) 3 . C . CO . CH 3 is 138*6, and if the volume of 0 " be equal
Digitized by
Google
28
LIQUID CHEMICAL COMPOUNDS.
to 2H, as is usual in such compounds, then the volume is equal
to that of trimethylethyl methane (CH 8 ) 8 . C—CH 2 —CH 8 .
Trimethyl methane.
CH,
CH,—i—CH,—CH,
(!h,
V» i 3 8 -6.
Pinakolin.
CH.
-<U
<1h,
138*6
The volume of the hydrocarbon, is thus, about 138 *6, a value
which is less than that of normal pentane by 139*9 - 138*6
= I *3. This value is,—
a = -1*3 = - 0*65 x 2,
or the contraction due to this tertiary group is twice that for an
iso group.
In some of the compounds at any rate, the contraction may
be due, in part at least, to an interaction of groups attached to
different carbon atoms. This is seen to be possible if we give to
trimethyl methane and pinakolin the following configurations—
CHo
V
j X CH s
CH*
CH 3
CH,
CHj CH 3
Trimethyl methane.
Pinakolin.
This possibility is lost sight of in the plane formulae, but the
solid models or diagrams to represent them, would show this.
This fact may be explained as follows :—
If we suppose, that, in the iso and tertiary groups, one CH 3
group is continuous with the main chain, then the other CH 8
group in the iso arrangement, and the other two CH 8 groups
in the tertiary, respectively occupy the $ positions, as shown in
the subjoined schemes.
I so Group.
CH 8 —CH—CHa—
ch 3
A = - 0*50
Tertiary Group.
CH,
CH;,—C—CH,—
c!h,
A = - 1*3 = 2 X - 0*65
The CH a groups, in effect, substitute hydrogen atoms from the
yS position, like chlorine or other atoms would, and consequently
the CH 8 groups occupy similar positions to the latter in the
Digitized by
Google
HYDROCARBONS AT THE BOILING-POINT
29
molecule. It consequently represents two propane residues,
joined together by their middle carbon atoms.
These formulae are similar to those upon which the “ Geneva
Commission” has based its system of nomenclature, and the
formulae suggested by it would seem to have theoretical justi¬
fication.
The formula for di-isopropyl, taking into consideration the
contraction is
Unsaturated Compounds.
In studying the hydrocarbons, it will be of great advantage
to consider the branching of the chain as equivalent to substitu¬
tions of the methyl group for hydrogen. Moreover, if more
than one substitution has taken place, the distribution may be
(а) Unsymmetrical CHX*. CH g
CX*: CH a
(б) Symmetrical CH,X. CH*X
CHX:CHX
CXj-.CX*
These features will affect both the molecular volumes and the
boiling-points of the compounds.
The following rules have been made out:—
CH S group
Cl
Unsymmetrical Distribution.
(CHj) 2 CH - AV - 0*5
(CH S ) S C- AV-ro
Cl* CH -
Cl 8 C - AV + i-5
Cl 4 C AV + 3*0
b.p. - 8*o
- 28*0
gradually diminishing differ¬
ences for Cl-H, showing
depression with continued
substitution.
Symmetrical Distribution.
CH 8 group - C(CH S ) • C(CHj) - AV - 1*5 to - 3*1 b.p. + 5*0
Cl - CHC 1 . CHC 1 - 31 + 23*6
- CHBr . CHBr - - 3*1 + 20*3
The following olefins are normal, or nearly so:—
Digitized by v^ooQle
30 LIQUID CHEMICAL COMPOUNDS.
Mono-Olefins.
Vm
V of sat.
compd.
less n x 7*
or 2 n Va
£ Amylene
CH 8 . CH 2 . CH : CH . CH 8
IIO*I
117*8
110*4
a Hexylene CH S . CH 2 . CH a . CH 2 . CH : CH 2
132*4
139*9
132*5
a Heptylene CH 8 . CH 2 . CH 2 . CH 2 . CH 2 . CH : CH 2
154*8
162*6
155*2
a Octylene
CH 8 . (CHJj. CH : CH 2
177*6
—
176*8
Di-Olefins.
Isoprene
CH 2 : CH. C(CH S ): CH a
103*6
—
—
Diallyl
CH a : CH. (CH 2 ) 2 . CH : CH a
125*1
I39’9
125*1
Aromatic unsaturated compound.
Styrolene
C 8 H 5 .CH:CH 2
131*4
139*3
131*9
The following olefins are abnormal:—
Pseudo butylene (dimethyl ethylene) symmetrical compound.
CH s .CH:CH.CHj b.p. + i°
b.p. of a compound - 5°
For CH S groups in a & pos. + 6
Calculated b.p. + i°
valerylene - <*13 0*653 (Chem. kal.) A = 0*00075.
d i 0*^53 + 12 x 0*00075 = 0*644 V m 86*9.
2 nV a — 96*0 - 7*4 = 88*6.
A = Vm - = 86*9 - 88*6 = - 1*7.
There is thus an increase in boiling-point for symmetrical
distribution (6°) and a contraction of - 1 *7.
Trimethyl ethylene (CH 8 ) 2 . C: CH(CH,)
b.p. 36° Vm 108*1 (Thorpe).
compare
C(CH3) a :CH(CH s )
A = - 2*0
with
CCl a : CHC1
A = - i‘5
for contractions
IIO*I
and for trimethyl ethylene 108*1
A = ~ 2*0
The contraction - 2*0 is connected with the symmetrical
distribution of the methyl groups.
Di-isopropyl.
CH(CH 8 ) 2 . CH(CH 8 ) 8
b.p. 58*0
b.p. n C 8 H 14 71 0
less two iso groups - 18 (2 x 9)
calculated 53
observed 58
A = + 5
b.p. of C 5 Hj h fi amylene 39 0
corr. for iso group - 8
3i
for a £ const. + 5
calculated b.p. 36
observed 36
Since the volume of
a or £ amylene is Vm
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT.
3i
There is thus an increase above the expected value. This
result is no doubt due to the interaction of neighbouring iso
groups. Also—
V m 136*8
2 nVa 139*9
A= - 3*1
The contraction for this compound is thus -3*1 below the
normal.
The following example from among the aromatic compounds
will help us to solve the problem of the constitution of these
compounds.
0 xylene
/ c \
b.p.
Vm
CEL C—
—och 3N
0 xylene
141-9
137*9
fl
\
1
1
p xylene
138-9
140*2
II
CH
—CHs '''
A
_+T 9
-2-3
There is thus an increase in the boiling-point of +3*9, and a
contraction of - 2*3 owing to the approximation of the methyl
groups.
A further study of the tri- and tetra-methyl benzenes shows
results which are similar.
Sym. dimethyl ethylene
AV
b.p.
CHj.CHjCH.CH,
Trimethyl ethylene
- 17
+ 6°
(CHj) 8 C:C(CH 8 )
Tetramethyl ethylene
- 2*0
+ 5 °
(CH JjC: C(CH,) f
Ditaopropyl
- 4*6
+ 14 °
(CHj) s CH. CH(CH,) 4
0 Xylene
- 3 *i
+ 5 °
C*H,(CHJ,
- 2-3
+ 3-9 from p xylene
+ 5*4 from ethyl benzene
The formula for 0 xylene would be
,CH=CH
CH
\
Digitized by v^ooQle
32
LIQUID CHEMICAL COMPOUNDS.
and by analogy
CH-
CH-
CU 3n
for symm. dimethyl ethylene compare
CH-CK
-ch 3 '
CH-
-Cl'
acetylene chlor¬
ide (q.v.)
Tetramethyl ethylene.
CH 3 Cl-
CH-
-CK
J
Chlor ethylene
chloride.
-Cl
'Cl-
-Cl'
Tetrachlor ethylene.
Diisopropyl.
CH 3 ch 3
CH CH
CH 3 ch 3
In all these cases where there is an augmentation of the
boiling-point and a diminution of the volume, there is spatial
proximity of the CH 3 groups. We therefore conclude that the
effect is due to an interaction of the groups.
A convincing piece of evidence is found in an examination
of the volumes of the di- and polyhalogen derivatives of the
paraffins.
The chlorine atom in propyl chloride (terminal) has a volume
of 21 *5 and in isopropyl chloride (central) a volume of 24*3. The
sum of the volumes of these two chlorine atoms is 45 *8.
Propylene chloride which has one chlorine atom in the a
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT .
33
position and one in the yS position has such a volume that the
sum of the two chlorine atoms is apparently 41*6. There is
thus a considerable diminution in volume caused by the presence
of the two chlorine atoms in the same molecule, one in the a and
and one in the # position.
Closed-Chain Hydrocarbons.
Aromatic Hydrocarbons and Ring Structure .
There is a constitutive effect in these compounds which is due
to ring structure, as already shown. Without going more fully
into the question of the cause of ring structure at present, we
simply calculate the values of SnV ai and express this correction
as a negative difference.
TABLE XIX.— Ring Hydrocarbons.
Substance.
.. .. . _
Vtn .
2»Va.
A,
Diamylene .
211*7
220*3
- 8*6
Benzene
• C 6 H 6
96*0
111*0
- 15*0
Toluene
. C„H 6 . CH,
118-3
133-2
“ 151
p Xylene
. C 6 H 4 (CH a ) 2
140’S
155*4
- 14*9
Ethyl benzene
• c,h 6 c,h 6
139*5
—
- 15*9
Mesitylene .
. C,H,(CH,) a
162*8
we
- 14*8
Propyl benzene .
• C e HjC,H 7
162*2
—
-15*4
Cymene
. C e H 4 (CH a )C,H 7
1849
199-8
- 14*9
Hexamethylene .
• c,h 12
ii6*3
133-2
- 16*9
Hexahydrotoluene
• C,H U CH,
141*8
155-4
- 13*6
Hexahydroxylene
. C e H 10 (CH 3 )j
164-3
177-6
- 13*3
Naphthalene
• CjoHg
147-2
177-6
- 30-4
Naphthyl hydride
• c 10 h 14
171*2
200*0
- 28*8
Anthracene.
• c 14 h 10
195*5
244*2
- 48-7
Simple Rings.
Hotnologues of Benzene .
Benzene C 6 H 6 Vtn 96*0 H 3*2.
Vw.
Toluene C 7 H 8 118*25 (Schiff)
C 6 H 6 - CH S
C 6 H 6 - 92*8
3
Digitized by v^ooQle
34
LIQUID CHEMICAL COMPOUNDS.
A for CH, 25-45 [CH,] = 25-45.
• p xylene C 8 H 10
C,H 4 (CH,), 140-5 (S.)
C,H 4 89-6
A for 2CH, 50-9 [CHJ = 25-45.
Ethyl benzene
CjHj . C,H, 139-28 (S.)
rC,HJ = 47-48
In paraffins 48-0
A - 0-5.
Mesitylene C,Hj,
C,H, (CH,), 162-67 (S.)
C,H, 86-4
A for 3CH, 76-27 [CH,] = 25-42
In paraffins 26-0 and larger
A - o-6.
Propyl benzene
C,H, . C,H 7 16216 (S.)
[C s H 7 ] 69-36
In paraffins 70*0
A - o*6.
The volume of CH 3 is evidently the same in all these com¬
pounds, and is slightly less than in those which have an open-
chain structure.
[CH,] (in the paraffins) = 25*9
[CH,] (in aromatic side chains) = 25*5
The groups C 2 H 5 and C 8 H 7 also show a contraction as
compared with similar residues in open-chain compounds. The
contraction is - o*6.
This is probably to be accounted for in a similar way to that
for the iso group.
Thus we have—
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT
35
The principle is not affected if we suppose that the paraffin
residues are directed along a line which cuts the rings at right
angles, or nearly so. This is equivalent to branching.
The volume of cymene may thus be synthesized:—
Cymene C 8 H 7 . C 6 H 4 . CH 3
C 6 H 4 8g-6
CH S 25*45
C 3 H 7 69*36
2»Va 184*41
Vm 184*88
p methyl ethyl benzene.
C 6 H 4 (CH 3 ) (C 2 H 6 )
c*h 4 89*6
CH 3 25*45
C 2 H 5 47 * 4 ^
5»V a 162*53
Vm 162*32
There are thus slight differences both + and - in the re¬
sults from which it would appear that small quantities of impur¬
ity might have been present in SchifFs material. For example
small quantities of the ortho modification in p methyl ethyl
benzene would account for the result. These compounds are
notoriously difficult to completely purify.
The Effect of Self-Affinity between the Side-chains of
Aromatic Compounds .
It has been shown that in the open-chain hydrocarbons self-
affinity is manifested by contiguous paraffinoid groups.
The same effect is well shown by the various aromatic com¬
pounds which possess two contiguous hydrocarbon side chains.
The data are due to Pinette (loc. cit.).
The xylenes C 8 H 10 .
Vm
A
C—CH,
c— ch 3
CH
/ \
✓ \
s \
CH CH
CH CH
CH C—CH.
1 II
1 11
1 II
CH CH
CH C—CH 3
CH C—CH S
\ /
\ /
\ /
C—CH,
CH
CH
para (i: 4)
meta (i: 3)
ortho 1(1:2)
140-25 - 0-25
140*00
I 37’95
Difference between o and pA = - 2*3
3 *
Digitized by v^ooQle
36
LIQUID CHEMICAL COMPOUNDS.
Strictly speaking two possible geometrical isomers of ortho
xylene should be accounted for, as perhaps also for the others.
They are:—
CH=CH
CH,
CH = CH
/
\l
/
CH
c
CH
1
CH-C
\
CH
-/I
I CH,
CH,
<!:h,
opposed. adjacent.
These isomers are constructed according to the well-known
fumaroid and malenoid types. If we examine the data carefully
we shall find that they agree with those which might be expected
to apply to the two forms.
Table XX.—The Volumes of the Xylenes.
Schiff. 34
Pinette. 25
Neubeck. 26
Para
v w .
140*5
b.p.
136-5
Vm.
140*25
b.p.
138*0
Vm.
140*77
b.p.
138-4
Meta
140*2
139*2
140*1
138-9
140*1
I39*i
Ortho
140*2
141*1
137*9
1
141*9
138*2
141*1
A between o and f
-0*3
-2*3
-2*07
It is observable that, as we pass from the para to the ortho
compound, there is, on the whole, an increase in the boiling-
point, and diminution in the volume. Comparing the values
given by the different observers, we find between the para and the
meta modifications, differences which are of a minor character.
It is to be observed in regard to the ortho compound, that the
boiling-points are similar, whilst the differences between the
volumes obtained by different observers, are far too great to be
ascribed to accidental causes. Pinette’s value is supported by
that of Neubeck. Schiff, on the other hand, was a chemist of
well-founded reputation for care and accuracy, so that his value is
at least entitled to equal weight.
It thus seems that the two sets of values refer to different
forms of the same compound. Moreover, the lower values differ
from that which might be expected by an amount equal to
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT,
37
about 2*5. This is a normal feature of ortho compounds, if we
consider the data available, and has been accounted for by sup¬
posing that there is an interaction between the groups owing to
residual affinity. The lower value would thus correspond to the
“Adjacent” orthoxylene.
CH=CH
/ \
CH C
%
CH —C
I CH,
CH,
V m = 138*2.
Schiffs larger volume is similar to that for paraxylene, thus
showing that the groups are independent and unable to interact.
This therefore corresponds to the
“Opposed” orthoxylene.
CH,
CH=CH I
/ \l
CH C
CH —C
SH.
Vm= 140*2.
On looking at the two formulae we see that there is every
reason to expect two forms. l
A study of other derivatives of benzene shows similar peculi¬
arities to those already noticed.
In the first place we must note the characteristics of “ ad¬
jacent 99 and “ opposed ” varieties of a compound. (See App. II.)
The esters of maleic and fumaric acid.
Maleate.
Methyl. b.p. 208°
Ethyl. 225
Maleic.
CH = CH
ioocH s £ooch 3
A Fumarate.
+ 16*0 192 0
+ 7*0 218
Fumaric.
COOCH,
I
CH = CH
ioocH,
Similar observations are made in the case of benzene de¬
rivatives. The ortho variety is usually some degrees higher in
boiling-point than the para, and this we believe to be due to a
conjunction of groups.
Digitized by v^ooQle
3 «
LIQUID CHEMICAL COMPOUNDS.
b.p.
p xylene
138°
m xylene
139°
0 xylene
142 0
J
b.p.
para
172 0
v
+ 4 °
Dichlorides.
meta
172 0
ortho
179 0
j
b.p.
para
219 0
V_
+ 7°
Dibromides.
meta
219 0
ortho
224 0
_ j
+ 5 °
The above and other data show that the meta is similar to
the para, but that the ortho is some degrees higher.
So we have shown that there is a difference in volume of
about - 2*0,
We, however, find apparent exceptions to the above rule.
The chlor toluenes.
para
b.p.
b.p.
para
163°
meta
- 13*0 150 + 6°
ortho
156°
- 7°
The brom toluenes.
para
meta
ortho
184°
183°
181 0
— 3
These anomalous results, where not due to association, may
very well be the result of stereo isomerism.
We inquire further into the question of the volumes of the
hydrocarbon di- derivatives of benzene.
The methyl ethyl benzenes.
-ch 3 \
S C-CH2-CH3''
C—ch 3
c-CH 3
c- -
/%
y\
CH CH
II 1
CH CH
11 I
CH C
II 1
<1h <1h
II I
CH C—CH 2 .CH 8
1 1
\ ✓
C-CHj.CH 8
b.p.
Vm
\ ✓
CH
para
162°
162*3
meta
159 °
1607
"V"
ortho
159 °
160*6
A - 17
Propyl benzene C 8 H 5 . CH 2 . CH 2 . CH S b.p. 158*6 Vm 162*2.
The difference in volume between the ortho and para methyl
ethyl benzenes is
A = 160*6 - 162*3 = - 17.
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT
39
From the fact that the volume of the para variety and the
volume of propyl benzene are almost identical, we suppose that
the methyl and the ethyl groups are independent, but from the
fact that there is a diminution of - i *7 in the case of the ortho
variety, we conclude that the groups interfere with each other.
They are on the same side of the ring and the compound is thus
of the adjacent type
The methyl propyl benzenes.
C-CH, C-CHg C-CHg
CH CH
CH CH
CH C—
CH 2 .CH 2 .
II 1
CH CH
II 1
CH C—CH,.CH,.CH,
A L
\s
W
\s
C—CH,.CH,.CH,
CH
CH
para
meta
ortho
b.p. 175*0 177
182*0
M
s
M
00
£
H
>*
—
Butyl benzene
C,H,CH,. CH,. CH,. CH,
b.p, 180
V*> 184*8
Iso butyl benzene
CH,
C,H,.CH,.CH<^
b.p. 167*5
Vm 1777
Secondary butyl benzene
CH,
C,H,. CH. CH,. CH,
b.p. 171 0
Vm 181*4
CH,
These last two compounds illustrate the rule that when un¬
saturated groups are attached to the benzene nucleus, large
contractions in volume are apparent.
The methyl isopropyl benzenes.
C-CH S C-CH, C-CHg
/ \
CH CH
II 1
CH CH
11 1
CH,
CH C—CH
II 1
CH CH CH ,
CH C—
CH<^
CH,
i:H CH
\S /
\S
C-CH<^
CHg
CH
3
CH
para
meta
ortho
CH,
/
\
CH,
b.p. 171 176 —
Vm 184*8 183*6 — t
A b.p. between para and meta 176 - 171 = + 5 0
A Vm between para and meta 183*6 - 184*8 = - i*a
It is remarkable that whilst there is no distinction between
the para methyl n propyl and methyl isopropyl benzenes as
regards volume, the meta normal compound shows a small com¬
pression and the meta iso compound shows a larger compression.
Digitized by v^ooQle
40
LIQUID CHEMICAL COMPOUNDS,
Diethyl benzene.
C-CH 2 • CH.
CH CH
L L
\ s
C-CHj.CHg
b.p. 179 0 Vm [185*4] 0*8707 C 0*520.
C 6 H 4 89*6
2C g H 5 96*0
185*6
Vm 185*4
b.p.
Tri substituted benzenes.
The trichlor benzenes.
1:3:5 1:2:4
208 213
: 2 :3
218
C—
CH CH
II I
CH 8 -C C—CH
W
CH
Mesitylene.
1:3:5
b.p. 164*5
+ 5 2 x + 5
The trimethyl benzenes.
CH s ,c-—ch 3
/ \
CH CH
CH
\
-CH 3 ',
c— —ch 3 )
Pseudocumene.
1:3:4
170
_/ V_
:ff\
-ch 3
-ch 3 ;
-CH S /
CH
Hemimellithene.
1:2:3
175
+ 5*5
+ 5
V*n 162*8
[160*9]
- 1*9
We see that there is an increase in temperature of + 5 0 for
each step of the change from mesitylene to hemimellithene, and
a contraction of - 1 *9 from mesitylene to pseudo cumene. We
cannot doubt that the volume of hemimellithene would be
162*8 - 2 x i*9= 159 0 at least.
The tetramethyl benzenes.
CH
/\
TT Vi
ch 3
•CH 3 '
CHs
.''CHs C \
v CHs——X j
CH
CHs-C,
-CHs'.
-CHs'
CH
Digitized by
Google
HYDROCARBONS AT THE BOILING-POINT
4i
Prehnitol.
204*0
Vm 184*4
180*0 = io*o = 2x5°
180*0 = 16*0 = 2x8°
180*0 = 24*0 = 3x8°
We thus see that two independent ortho groups occasion a
rise in the boiling-point of twice that for a single ortho group.
Two contiguous ortho groups occasion a rise of twice 8° and
three contiguous ortho groups a rise of thrice 8°.
The two contiguous ortho groups of isodurene cause a con¬
traction A = 184-4- 181*8 = - 2*6 = 2 x - 1*3.
In reviewing the work done in the last section we note that
whenever two methyl groups
; are
in association there are
augmentations of + 5°
in the
boiling-point and contractions of
- 17.
Examples.
A.
A.
p xylene
140*3
- 2*4
one ortho
0 xylene
137*9
p methyl ethyl
162*3
- 1*7
one ortho
0 methyl ethyl
160*6
1:3:5 trimethyl
162*8
- i *9
one ortho
1:3:4 trimethyl
160*9
1:3:5 dimethyl ethyl
184*8
- i*6
one ortho
1:3:4 dimethyl ethyl
183*2
Tetramethyl calculated
1:2:3 :5 tetramethyl
185*2
181*8
- 17 X
2 two ortho groups (contiguous)
/V CH3 ^
ch t:-ch 3 ;
CH 3 v
/
Durene.
b p. 190°
Vm —
C--CHa
Isodurene.
196*0
[181*8]
Butyl benzene.
C 6 H 5 . CH 2 . CH a . CH 2 . CH 3 b.p. 180*0
A b.p. between butyl benzene and durene 190*0 -
A b.p. between butyl benzene and isodurene 196*0 -
A b.p. between butyl benzene and prehnitol 204*0 -
It is thus concluded that whenever we have two methyl
groups in conjunction there are contractions of -1*7 or there¬
abouts, and as often as this is repeated in a single molecule, we
find this contraction.
From this it follows that those compounds which have been
examined are the “adjacent” varieties and the formulae of these
would be
Digitized by v^ooQle
42
LIQUID CHEMICAL COMPOUNDS.
o xylene. o methyl ethyl. 1:3:4 trimethyl. 1:3:4 dimethyl ethyl.
If the substituents be methyl groups, they are unable to
interact when in the meta position relatively to each other, but
there are indications that when the chain is lengthened they may
do so.
V*». A.
p methyl ethyl benzene 162*3 _ i*(
m methyl ethyl benzene 1607
p methyl isopropyl 184*8
m methyl isopropyl 183*6
one meta
one meta
1:2:5 dimethyl ethyl
1:2:4 dimethyl ethyl
184*8
181*6
- i*6 x 2 one ortho and one meta
The formulae are thus
m methyl ethyl. m methyl isopropyl. 1:2:4 dimethyl ethyl.
There are exceptions to this, but the volumes are in all cases
normal.
V*. A.
p methyl propyl 184*8
m methyl propyl 184*1 ” ° ^
1:3:5 dimethyl ethyl 184*8 (1)
1:3:5 dimethyl ethyl 183*9 (2) °’ 9
This, so far as we can see, can only be explained by placing
one group in opposition to the others.
m methyl propyl. 1:3:5 dimethyl ethyl
benzene.
We assume that the opposed varieties of the different com¬
pounds would have normal volumes because the interaction of
groups is impossible.
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT.
43
All this seems to agree with our explanation of the two
values for o xylene.
CH*
<7
ch 3
Vm 140*2 (Schiff)
140*6
Complex Aromatic Rings.
(a) Condensed.
The benzene nucleus is the only one of the type, which con¬
tains a single ring, for which data are available. The following
is a series of such hydrocarbons. See Le Bas (loc. cit.) and
ref.
A ' H
chJi
\
(1)
CH
XH—CH
CH—i
CH=CH
/
/
1 1
CH,
CH
CH*CH
\
CH—CH
CH—
(2)
(3)
(4)
\
CH
✓
Derivatives of all the above hydrocarbons are known except
(1), but, as before stated, only (4) has been prepared.
We now proceed to study a series of more complex hydro¬
carbons, which are wholly or in part of the type just mentioned.
The series is made more complete by the use of the formula
described in the appendix (q.v.).
\ /\/
CH CH*
Hydrindene. C 9 H ]0
dm 0*957 b.p. 176° C =5 0*46
V m . 144*0 (by formula)
2 nVa. 170*2 (46 x 3*7)
A for ring structure = -26*2
One six-membered ring -15*0
One five-membered ring -12*0 (vide ante)
Total contraction = - 27*0
Digitized by v^ooQle
44
LIQUID CHEMICAL COMPOUNDS .
CH CH
s \/ %
CH C CH
in c 1 <S:h
\ /\ ^
CH CH
Naphthalene Ci 0 H 8 .
V« 1472
177*6 (48 x 3*7)
A for ring structure = - 30*4
Since the contraction for a single six-membered ring is
- 15-0, the contraction for two such rings is - 30*0. A con¬
traction of this magnitude is seen to apply to naphthalene, a
compound containing a condensed ring system, which is evidently
equivalent to two separate six-membered rings.
CH ch 2
s \/ \
CH 2
Ah,
\ /\ /
CH CH,
Naphthyl hydride C^H^.
Vm 171*2
2»Va 200*0 (54 x 3*7)
A for ring = - 28*8
If we compare naphthyl hydride with naphthalene, we get the
following results:—
[C 10 H 14 ] *=* 171*2
[C 10 H 8 ] - 147*2
A for 6[H] = 24*0
Again, comparing toluene and hexahydrotoluene, we find—
[C 7 H 14 ] - 141*8
[C 7 H 8 ] = 118*3
A for 6 [H] = 23*5
It is thus seen, that, on reducing a substituted ring of the benzene
extra hydrogen atoms possess volumes which are nearer 4*0 than
to 37.
The result is, that the apparent contraction for such a re¬
duced ring, is less than the normal, i.e., by an account equal
to 6 (4*0 - 37) = 1 '8. Thus the contraction for the nucleus really
is - 28*8 - i*8 = - 30*6. For naphthalene - 30*4.
Digitized by v^ooQle
45
HYDROCARBONS AT THE BOILING-POINT
Acenaphtenc C ]2 H 10 .
CH
CH
0 1
L
\
\/
c
\
c
/
c
CH
-CH„
m.p. 103°, b.p. 277 0 d m.p. 1*03.
By formula db.p. 0*903 if c = 0*46.
V m = I 70*5
5 w V a = 214*6
A for ring = - 44* 1
Two six-membered rings 2 x 15 - 30*0
One five-membcred rings - 12*0
Total contraction =* - 42*0
Anthracene C 14 H 10 .
CH
CH CH
c4
i„
\
\/
- c
CH
C
/\
\/
C
V
CH
CH
C C
/\ ✓
CH
H
V» 195*5
2 n Va 244*2
A for ring = “ 48*7
Two six-membered rings - 3 °‘°
Two four-membered rings - 17*0 (2 x 8*5)
Total contraction - 47*0
Three six-membered rings would show a contraction of
3 x - 15-0 = - 45 0, from which it follows that the above
formula is the correct one.
Phenanthrene C 14 H 10
CH=CH CH=CH
/ \ / \
(1) CH C-C
\ S \
CH—C C
\ /
CH=CH
s
-CH
CH=CH
/ \
CH (2) CH
\„-
/
-C—
CH=CH
/ \
=C
\
—c-
\ /
CH=CH
CH
CH
Vw 196*0
2»Va 244*2
A = - 48*2
The contraction for phenanthrene is similar to that for anthra¬
cene. It is thus probable that they have similar structure. If
this were the case) phenanthrene would have a formula similar to
(2) and not like that of (i).
Digitized by v^ooQle
46 LIQUID CHEMICAL COMPOUNDS.
This structure would occasion a contraction of - 48*0.
Two single six-membered rings 2 x 15 = - 30*0
Two four-membered rings 2 x 8*5 = - 17*0
A = - 47*0
The first formula would show a contraction equal to 3 six-
membered rings 3 x -15 0= - 45*0
(b) Separated Rings.
No data are available for compounds of this type, but with
the help of the formula, we may find out how these rings affect
the volume.
Diphenyl C ia H 10 .
b.p. 254 0 d^ 0-9961
C = 0*46 Vm 180-0
2 C 6 H 6 192-0 2 x 96-0
less 2 H 6-4
185-6
V w =i8o-o (by formula)
A = - 5*6
There is reason to suppose that there is a special constitutive
effect associated with the presence of two conjugated unsaturated
rings. Accurate data, however, are alone capable of showing
this definitely.
m.p.
7°'5
&
CH CH
/ \
CH CH
I II II I
CH CH CH CH
\ / \ /
c-c
The Polymethylenes C n H sn .
The aromatic compounds, when reduced, give a series of
saturated compounds called the polymethylenes. They are
isomeric with the open-chain olefins, and are distinguished from
them by the possession of ring structure. Only two or three
compounds of this character have been directly investigated, viz.
diamylene C 10 H 20 , hexamethylene QH 12 and its homologues
hexahydrotoluene C 6 H U (CH 8 ), hexahydroxylene C 8 H 10 (CH 8 ) 2 .
The values obtained for the others must consequently be calcu¬
lated values. C = 0*46, which is a number usually employed
when unsubstituted ring compounds are dealt with.
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT
47
TABLE XXI.— The Polymethylenes.
Olefins.
Vm.
A for Ring.
Vm.
Polymethylenes.
Propylene CsH 6
Butylene C 4 H 8
Pentylene C 5 H 10
Hexylene CgH^
Heptylene C 7 H 14
Octylene C 8 Hj 6
66*6
88*8
110*0
132*5
154*8
177*6
-6*4
-8*8
- 12*0
- 16*0
-20*8
! - 26*4
60*2
8o*o
98*0
116*5
i34*o
151*2
Trimethylene (CH 2 ) S
Tetramethylene (CH 2 ) 4
Pentamethylene (CH a i s
ft examethylene (CH a L
Heptamethylene (CH a ) 7
Octomethylene (CH 2 ) 8
The diagram on next page shows how the contraction for ring
structure varies with the size of the ring.
Confirmative evidence in favour of the above contractions for
single-ring structure is found in the following table. Since it is
derived from the consideration of derivatives of the simple hydro¬
carbons, details concerning the derivation of 2«V a must be looked
for elsewhere.
TABLE XXII.— The Magnitudes op the Contractions due to Ring
Structure.
For For
Polymethylenes. Vm. ring ring Vm.
structure, structure.
60*2 - 6*4
Trimethylene—
CH 2
L> ch '
Tetramethylene—
CH a — ch 2
I I 8o*o -8*8
CH 2 —CH 2
Pentamethylene—
CH a —CH a
!h„—CH.
\
/
CH 2 98*0 -12*0
H examethylene—
^ch 2 -ch 2 ^
ch 2
V
CH 2 116*5 -i6*o
- 5*2
■8-5
12*4
-15*0
Other compounds.
Epichlorohydrin
1 87*5 _a
92-2 2»Va | \ Q
ch/^
Vfn. Diamylene.
2117 C 3 H 7 —CH—CH a
k 220*2 juVd d/Hj—i/H—C 8 H 7
Vm. Thiophene.
CH=CH
849 , \_
97*3 2#V* c H=CH /
Vm.
96*0
Benzene.
CH—CH
CH
iii*o 2 n Va \
w—*
CH
CH—CH/
Digitized by v^ooQle
48
LIQUID CHEMICAL COMPOUNDS.
It cannot be doubted, on the one hand, that the above-men¬
tioned contractions are the true ones, if we allow for the possi-
Numbers of Members in Ring.
Fig. 2.
bility of a slight variation one way or the other, and on the
other hand, that the volumes calculated for the polymethylenes
are approximately correct
Digitized by
Google
HYDROCARBONS AT THE BOILING-POINT.
49
The most interesting fact connected with these results is the
one already noticed at the commencement of this chapter, viz.
that the contraction for ring structure varies with the size of the
ring . This is the result which we should expect, if the contrac¬
tion is additive in nature, that is, made up from the sum of small
deviations of the individual atomic volumes from the normal. It
is, however, remarkable, that those attempts which have been
made, to elucidate this question, by means of other physical
properties which show a marked effect for ring structure, give
results which are opposed to the conclusions arrived at here.
The view taken by Sir Wm. Perkin, in his study of molecular
magnetic rotations, 27 is, that the magnitude of the constitutive
effect is constant , and thus not related to the size of the ring.
Independently of the question of the probable correctness of
this latter view, a study of the question has led us to conclude
that it is probably based on a wrong reading of the data. It
is remarkable that Perkin at first calculated results which
favoured the view, that the magnitude of the effect due to ring-
formation varies with the size of the ring; but because these
results appeared to him to be anomalous, he discarded this method
of calculation for another one which seemed more satisfactory.
TABLE XXIII. —Comparison of two Methods of Calculating the Effects
of Ring formation on the Molecular Magnetic Rotations of Organic
Compounds.
No. of
Members
in Ring.
Substance.
M
For Ring
Structure.
For Ring
Structure.
M.
Substance.
f Butyric Acid
4‘47 2
4-740
Formic Acid + 3CH 2 .
3
-[ Trimethylene . }
-0*331
-0*599
f Trimethylene.
[ Carboxylic Acid j
4-141
4141
{ Carboxylic Acid.
f Valerianic Acid.
5*513
5*594
Acetic Acid + 3 CH S .
4
-J Tetramethylene \
-0-465
-0-546
/ Tetramethylene.
[ Carboxylic Acid j
5-048
5-048
1 Carboxylic Acid.
f Hexylic Acid .
6-530
6-531
Propionic Acid + 3CH 2 .
5
-j Pentamethylene \
-0-639
- 0*640
/ Pentamethylene.
\ Carboxylic Acid J
5-891
5-891
i Carboxylic Acid.
f n Hexane .
6-646
8-582
Valerianic + 3CH 2 .
6
\ \
-0*982
-0*607
j Methyl Hexamethylene.
l Cyclohexane . j
5-664
7*975
1 Carboxylic Acid.
4
Digitized by
Google
5o
LIQUID CHEMICAL COMPOUNDS.
The magnitude of the effects for ring formation on the left-
hand side of the above table, favour the conclusion arrived at
from a study of molecular volumes, whilst those on the left-hand
side, which were adopted by Perkin, give a different result
It is not difficult to show, by making a summation of the
atomic values, or by means of a calculation of the series-differ¬
ences S, that formic and acetic acids are anomalous, and not
only these, but acetone and other compounds employed by
Perkin in calculating his results. The average value of S the
Series constant for the acids is 0*393, i.e. for O a .
TABLE XXIV.— Acids.
Substance.
M.
A.
S.
'Zntn.
Difference.
Formic Acid
Acetic „
Propionic „
Butyric ,,
Pelargonic ,,
1-671
2*525
3- 462
4- 472
9590
0-854
0*937
i-oio
5 x 1-023
(0-648)
(o* 479 )
o *393
0-380
0-383
1- 410
2- 419
3- 462
4- 485
9*600
- 0*261
- o*i 06
± 0
+ 0-013
+ 0*010
The Ketones.
Acetone.
Methyl Propyll
Ketone / *
3-514
5-499
2 x 0*992
(o- 445 )
0-384
3*444 |
5-508
- 0-091
+ 0-009
These first and second members of the respective series are
not strictly analogous to the succeeding members as regards
their molecular magnetic rotations, any more than they are for
other physical properties, and thus they cannot be used to
calculate the effects due to ring-formation. The values of M are
relatively high for formic, acetic acid, etc., so that the calculated
effects, due to ring structure, are thereby increased. This result
makes it appear that the effect is independent of the size of the
ring, a conclusion which is seen to be incorrect
Perkin argued that because tri methylene carboxylic acid
is the first member of the series, it should possess an augmented
value similar to formic acid. This is not necessarily the case,
because the former compound is differently constituted, since it
contains a ring, and in any case, the amount of the augmentation
differs considerably, even for different open chain compounds.
Trimethylene carboxylic acid is sufficiently complex to exclude
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT.
5i
the idea of such an augmentation, and it is of nearly the same
complexity as butyric acid, with which it may suitably be com¬
pared. We believe that the position of a compound in a series
has no significance, apart from complexity.
It must also be observed that magnetic rotary power is a
physical property which is much more sensitive to constitutive
influences than molecular volumes, so that even slight changes in
structure might considerably affect the values. An apparent
difference in the observed and calculated results, and which is
relied upon to indicate the extent of some constitutive influence,
may consequently be the algebraical sum of several effects. This
would vitiate to some extent the calculated effects due to ring
structure determined by this property, and probably does so.
In molecular volumes , on the other hand, we choose data
which are only affected by the constitutive influence in question
—ring structure, the others being negligible by comparison.
Moreover, the corrections for this structure are very con¬
siderable.
No physical property is probably so well adapted to the elucida¬
tion of the ring structure of chemical compounds as that of
molecular volumes .
The Naphtenes .
These subjects are specially interesting, chiefly owing to the
fact that they occur in many mineral oils and other natural pro¬
ducts. They do not, however, differ in character from the com¬
pounds of the preceding group. The data have been taken
chiefly from Richter’s Organ . Chem . (Smith). Eng. edit. vol. ii.>
p. 292.
Hexahydrotoluene C 7 H 14 (Heptanaphtene).
CH a
/
CH,
Ah,
CH,
CH—CH.
\ CH /
Vf« = 141*8 (Lossen and Zander).
[C 6 H 5 ] = 92*8
6[H] = 24*0
[CH3] = 2 SS
%Na = I42‘3
For ring formation. Hexahydrotoluene C 7 H 14 141*8
Heptylene „ 154*8
A for ring = -13*0
4 *
Digitized by v^ooQle
52
LIQUID CHEMICAL COMPOUNDS,
This value does not represent the real contraction for ring
structure, since in these compounds the six extra hydrogens
take up larger volumes (■ v.a .) than is usual with ring compounds.
p Hexahydroxylene C 8 H 16 .
CH—CH S
/ \
CH, CH,
Vm = 165*0 (L. and Z.).
[C 6 HJ = 89*6
2CCH3] = 51*0
1 1
CH, CH,
6[H] = 24*0
—
N CH-ch 3
5»Va = 164*6
For ring formation. Hexahydroxylene C 8 H 16 165*0
Octylene „ 177*6
A
-124
Hexahydro m xylene C 8 H 16 (octonaphtene).
The value of C for the preceding compound is C = 0*506.
CH--CHj,
/ \ b.p. 118 0 dn 0*7814
CH, CH,
| | db.p. 0*6782 Vm 165*1 by formula
CH a CH—CH*
\ / 2«Va 164*6
XH,
According to theory, there should be little difference between
the volumes of this compound and that of the preceding one.
This is found to be the case.
Hexahydrocumene C 9 H 18 .
CH 3
/
CH—CH
/ \ \
CH, CH, CH,
CH, ^H,
\ /
CH,
b.p. 148° djo 0*7870
C = 0*506 + 0*031 = 0*537
d b.p. 0*6766 V« = 186*2
[C 6 HJ = 92*8
= 70-0 (19 x 37)
= 24*0
2»Va = 186*8
Vm — 186*2
[C s H 7 ]
[6H]
Or from cymene.
Cymene C 6 H 4 (CH 8 ) (C 3 H 7 ) 184*5
less CH S - 25*5
159*0
plus H 3*2
162*2
Six extra hydrogen atoms 24*0
2»Va = 186*2
V w = 186*2
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT .
Hexahydropseudocumene C 9 H 18 (Nononaphtene).
b.p. 135 0 d ia 07812 (Konowaloff).
CH —CH S
C
o *537
/ \
CH, CH—CH,
1 t
d b.p.
0*6738
CH 2 CH a
v,«
187*0
\ /
CH CH a
Mesitylene C 9 H 15I
162*8
6H
24*0
186*8
Vm
187*0
53
It is necessary to note that although the compound contains
two CH 3 groups in the ortho position, an arrangement which is
generally accompanied by a contraction of about - 2*5, this does
not appear from a consideration of the calculated volumes.
It might be explained by assuming that the particular
specimen examined was a compound with thetrans or “ opposed ”
configuration.
11
ch 3
CH3 CH3
Opposed ” Hexahydropseudocumene.
We do not, however, wish to lay much stress on this con¬
clusion, since the value is one which has not been found experi¬
mentally. The volume of the adjacent form would be
186*5 - 2*5 = 184*0.
QH3
ch 3
** Adjacent * ’ Hexahydropseudocumene.
A General Study of the Terpenes.
The data have been taken from Die Atherische Ole,
Semmler, vol. i., 1906.
(a) The Hemi-terpenes C 6 H 8 .
^C-CH = CH S
CHj/
Isoprene.
V« 103*9 (Buff).
2»Va 103*6 (28 x 37).
Digitized by v^ooQle
54
LIQUID CHEMICAL COMPOUNDS.
Isoprene is an open-chain diolefin. The volume of valerylene
C 6 H 8 is 104 2.
(b) The Olefin Terpenes (C 5 H 8 ) 2 .
This is a little understood class of compounds. They are
probably due to the polymerization of two molecules of one or
other of the hemi-terpenes, at least in theory.
The difficulty in calculating the values of these compounds is
to know what value for C applies to them.
From di-isoamyl iCjqH^ d^g 07358 b.p. 159*4 (Schiff).
db.p. 0*6126.
By calculation we find a value of 0 58 for C. This gives
numbers which are about 5 o units lower than those calculated
by the method of summation.
Their theoretical volumes are:—
C 10 H M (diisoamyl 232
less 6 H - 23
= 209 or 207*8 by the usual method.
TABLE XXV.— The Volumes of the Olefin Terpenes.
(Calculated by Formula.)
Compound C 10 H 16 .
b.p.
d /.
Vm by
Formula
2»V*.
A.
Myrcene
172*0
0*803IjQ.
203*0
207*8
-4-8
Alloocimene .
188*0
0-8172*,.
202*4
»«
~ 5*4
Anhydrogeraniol
172*6
0-8232,*.,
198*3
»»
- 9*5
Linaloolene C J0 H 18
165*8
0-7882*,
208 *9
214*6
- 57
It is difficult to completely account for these numbers by
means of the ordinary open-chain formulae. Since the values are
calculated, it follows that we must wait for directly determined
data before we can draw certain conclusions.
On the other hand, the nature and origin of the compounds
by no means excludes, in some cases at any rate, ring structure.
Take, for instance, myrcene.
The ordinary formula requires little modification to produce
the following:—
CH 8
\C : CH . CH 2 . CH a —C—CH
CH »/ ji 1
CHjCHj
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT
55
This would cause a contraction of about - 3*o.
From this there is an easy transition to
CH,^
CH S /
C : CH . CHj|. CH a . C=CH
I I
CHjt-CHj
which would cause a contraction of - 6*o.
If we study the formation of the compounds by the dehydra¬
tion of certain alcohols, we also see the possibility of ring struc¬
ture.
CH3 >
ch y
: CH.
Geraniol ClflHiaO.
CH 2 . CHj. C (CHj) : CH. CH 2 OH
Triolefin (C I 0 H 1B ). Ring compound (C 10 H 16 ).
CH».
>C : CH . CH, . CH. . C(CH,) : C : CH.
CH,/
CHj\
\c
CH,/
Anhydrogeraniol
: CH . CH . CH—C(CH.)
U
Both of the above compounds are possible, according as the
neighbouring or the fourth hydrocarbon group provides the
hydrogen atom which unites with the hydroxyl group to form
water.
I Linalool C l 0 H 18 O.
ch 3 .
\C : CH . CHj . CH, . C(CH 3 ) OH . CH : CH,.
CH 3 /
Triolefin (C 10 H !6 ).
CH,.
ch ^>C : CH . CH, . CH, . C(CH 3 ) : C : CH,
Ring Compound (C 10 H 16 ).
CH 3\
)C : CH . CH,. CH, . C(CH 3 )—CH
ch 3 / 9 \V
CH
The above triolefin is similar to the preceding one—that is
a similar dehydration of geraniol and linalool produces anhy¬
drogeraniol, if the open-chain formula properly describes it
Linaloolene has two hydrogen atoms more than the com¬
pounds of formula C^H^.
We think that it will be found that myrcene, alloocimene and
linaloolene are the ordinary open-chain triolefins, but that an¬
hydrogeraniol, or the compound which is called such, is in reality
Digitized by
Google
56
LIQUID CHEMICAL COMPOUNDS .
CH,v
>C : CH . CH a . CH — C (CH,)
CH,/ I ||
CH a —CH
With this conclusion the contraction found in the table is in
agreement
Moreover, the molecular refraction of anhydrogeraniol differs
from that of myrcene.
Myrcene. Anhydrogeraniol.
d l6 0*8025 /ad 1*4673 rfao 0*8232 /ad 1*4835
M * 46*9 M = 47*2
The values of A for myrcene allocimene and linalloolene
can, at least in part, be thus explained.
(a) The group
ch 3 \
)C = CH -
CH,/
is responsible for a contraction of — 2 *4 as in amylene.
( b ) The second iso group might cause a contraction of 09
as in compounds of complexity C 8 .
Thus a - - 2*4 - o*9 - - 3*2
1 nUS 2nVa - 207 8 - 3 2 = 204*6
which is not very different from 203*0, which might be in error
by a unit or so.
(V) The Menthan or Terpan Terpenes (CgHg^.
These compounds, which occur in many essential oils, are
probably formed by the polymerization of isoprene or some other
hemiterpene, with the formation of a ring. Thus:—
CH 2 CH a C^ CH,. X CH^-CR,
C—CH C—CH, -*■ C—CH C— CH,
CH,^ CH,=CH ^ CH,^ ^CH,—CH ^
Two molecules of isoprene. Carvene.
The volume of carvene is 190*3 (Schiff).
The observed value for cymene V*» 184*6
5nVa 184*8
Cymene
less 2 [H]
Adft]
184*6
- 7*4 for unsaturation
- of C,H 5 group.
177*2
12*8 if the four nuclear H atoms
- are similar to those of benzene.
190 o for carvene,
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT
57
If, on the other hand, the four hydrogen atoms are similar
to those in hexahydrotoluene [H] = 4*0, then the volume would
be 193'2, which is about 3 units too large.
The four extra hydrogen atoms are thus equivalent to four
hydrogen atoms in benzene.
The calculation verifies the formula for carvene.
CH a CH a —CH a
\ / \
C—CH C—CH 3
/ \ ^
CH 3 CH a —CH
We may now study the two phellandrenes.
CH 3 ch 3 CH 3 ch 3
\ /
CH
\ /
CH
i„
/ \
/ \
CH a CH a Phellandrene.
CH a CH |
1 11
1 «
CH CH
CH, CH
\ /
\ /
C
C
1
II
CH S
CH,
a Phellandrene b.p. 170® C.
d 15-5 0-846
& »» »» »!
0*848
By formula : a form, 190*2 ; $ form, 189*7.
Phellandrene.
C = 0*523
Calculating the volume of the a form by summation, we find
[C 10 H 14 ] = 184-6
2[H] = 6*4
2 nVa = 191*0 for a phellandrene.
It has been shown that the hydrogen atoms in the nucleus, or
those which partially reduce it, may
(a) resemble those of benzene, H = 3*2 V« 190*3
(b) „ „ hexahydrotoluene H = 4*0 Vm 193*0.
Carvene has been shown to answer to the first condition, and
apparently a phellandrene also.
Limonene and dipentene on the other hand answer to the
second condition.
Limonene b.p. 177*8 d, 5 . 5 0*848 C = *523 V m 193*1
Digitized by v^ooQle
58
LIQUID CHEMICAL COMPOUNDS.
TABLE XXVI.— The Calculated Volumes op the Terpenes.
Terpenes C^H^.
b.p.
d 16 - 5 *
db.p.
V w (calc.).
/*•
f Limonene
y Dipentene
Carvene
Sylvestrene .
- a Phellandrene
0
Terpinene
i 77 °-i 78
178
175
176
170°
170°
180 0
0*848
0*848
0-853*)
0-85115-5
C846
0*848
O-849
0-7043
0-7043
07159
07165
0-7150
0-7169
0*7134
i 93 *i
i 93 *i
190-0
189*8
190*2
189-7
190*6
1-4746
1-4740
by obs. 190-3
i *4747
i *4732
i *4759
1*4846
Whether these distinctions are valid or not, remains at present
uncertain, but in any case the various menthan terpenes possess a
range of volume of only 1897 - 190*6 or at most 1897 - I 93 'i.
This of course indicates that their structures are so little
different, that the molecular volumes are not very much affected
by such differences as do occur. These are mostly indicated by
the positions of the two ethenoid linkages.
One or two other terpenes of a somewhat different type re¬
main to be noticed.
Thujen b.p. 151 d 15 . 8 0-851.
If C as 0*523 d b.p. 0-7281 V m = 186-5.
The volume of this compound is considerably smaller than
the terpenes just enumerated.
The formulae which have been ascribed to a and ft tanacetene
are:—
a Tanacetene.
CH3 CHs
OH
CH*
Tanacetene.
CH3^ ^OH3
CH
CH OH
\/
cb 3
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT
59
By summation, we find
Ci 0 H |4 199*8 (54 X 37)
2H 6*4
'ZnVa = 206*2
Vm = 186*5 (by formula)
A = - 197 for ring.
This contraction is somewhat larger than for a single six-
membered ring, viz. - 15*0.
Two four-membered rings 2 x - 8*5 = - 17*0
One three and one five-membered ring - 6*3 - 12*0 = - 18*3.
The observed contraction is nearer the latter than the former.
In any case it is not compatible with the idea of a single six-
membered ring, so that tanacetene must be considered to be
possessed of a bi-cyclic ring.
A further investigation of bi-cyclic compounds of this type is
not desirable at present, since the value C = 0*523 which has
been applied to ordinary terpenes may not be quite true for
these, and we have no means of ascertaining the true value.
Direct data for one such compound would furnish the neces¬
sary material for such an investigation, but it is at present
wanting.
One point of interest exists, which is, that the contraction for
a single six-membered ring does not differ much from those
marked by what we may call cross-linking . This term is used
to distinguish these from bridged rings . Thus we may men¬
tion :—
n a = - 16*o
(H+4 A = 2 x - 8*5 = - 17*0
riHy + r A = 6*5 + 12*1- 18*5
The small difference between the contractions for the two
classes of compounds makes it difficult to distinguish between
them, since a slight difference of this magnitude might be due to
small additional constitutive effects.
The following is a list of the reduction products of cymene :—
Digitized by v^ooQle
6o
LIQUID CHEMICAL COMPOUNDS.
TABLE XXVII.
Compound.
Vm.
Apparent
effect of ring
formation.
3 »V«.
C.
Cymene C 10 H 14
184-6
- 15*0
184*8
0*523
Carvene C 10 H 16
190-3
- 169
190*3
»»
Menthene C 10 H 18
[i 9 8-8]
- 15*8
197*4
»»
Hexahydrocymene C^H^
[808-3]
- 13*0
208*6
»»
The six extra hydrogen atoms in the fully-saturated com¬
pound hexahydrocymene possess volumes of [H] = 4*0.
The hydrogen atoms in the nucleus of all the other com¬
pounds have been regarded as similar in volume to those of
benzene [H] = 3*2.
(c) Catnphane Terpenes C 10 H 16 .
These terpenes belong to a second class of ring compounds
which are distinguished by the possession of bridged rings. None
of these have had their molecular volumes directly determined
except pinene, but two derivatives— camphor C 10 H u O and
borneol C 10 H 180 —have been determined by Kehrmann 28 . These
data are sufficient to enable us to ascertain the nature of the
rings which are characteristic of this class of compound.
Camphor.
CH 2 - CH -CH 2
CH3-C-CH3
CH 2 -C-C:0
1
CH3
V« 188-6.
We find by summation—
CujHw 56 X 3-7 = 208*2
: O ii*o
2nV a 219*2
Vm 188*6
A = - 30*6
Borneol.
CH 2 -CH-CH 2
CH3-C-CH3
CH 2 - C-CH (OH)
I
CH3—
v m 190*5.
C 10 H 18 58 x 3*7 = 214*6
•O. T4
2nVa 222*0
Vm 190*5
A= - 31-5
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT
61
The ring structure of these compounds is thus found to give
rise to a contraction of about - 30, which is double that for a
single six-membered ring.
In consequence of the absence of data we shall be under the
necessity of calculating the molecular volumes by means of the
formula. The value of C is doubtless 0*460, since this value has
been found to be appropriate for a large number of ring com¬
pounds which do not possess side chains of more than one car¬
bon or equivalent atom.
Camphene C 10 H 16 m.p. 48° b.p. 160°
d^ 0*8481 0 = 0*46
d b.p. 0*7578 V m = 179*5
Ci 0 H 16 207*2
V w 179*5
A = -27*7 for ring.
This difference represents the value of the contraction for
ring structure, and is much larger than for a single six-membered
ring.
We thus conclude that the ring structure is similar to that
which has been found for camphor and borneol.
For two five-membered rings A = 2 x -12 = -24
For two six-membered rings A = 2 x -15 = -30
Camphene C 10 H 1(J may theoretically be supposed to be formed
as follows:—
CH S —CH—CH S
CH a —ili—CH 3
iH 2 — CH(CHJ— CH S
CH a -
-CH-
-CHo
CH,
A
Ah — ch
II
H 2 —CH(CH a )—CH
Ah,
Ah,
By analogy with camphor and borneol, a compound with the
above structure should possess a contraction of about - 30*0.
The calculated value is about - 28*0.
As before indicated, two five-membered rings produce a con¬
traction of - 24*0, which is a number considerably below - 30*0.
Digitized by v^ooQle
62
LIQUID CHEMICAL COMPOUNDS .
It is difficult to account for this, on the basis of the plane for¬
mula, but remembering the tetrahedral arrangement of the car¬
bon valencies, we may substitute the following one, which has
also been proposed by O. M. Foster, 28 for bromnitro camphane
and other compounds of this type.
This formula shows the two halves of the camphor molecule
forming an angle with each other like a partially opened book
set on end.
The carbon groups shown by the dotted line, together with
the corresponding ones above, are in a position to influence one
another so as to form a potential four-membered ring—
CHj CH 2
!h 2 .Jh(OH)
The additional contraction for this, would have to be added
to - 24*0, for the two five-membered rings, and so the observed
contractions would be partly accounted for.
Pinene C 10 H lfl .
CH 2 —CH-
/ \CH a
CH — C-
-CH,
CH
b.p. 156° d b.p. 07421 (Schiff). 49 *
V w = 183*2
. By formula—
cLjo 0*858 C = 0*46
d b.p. 07487
Vtn 1817
Since [C 10 H 16 ] = 207*2
Contraction from observed value A = 207*2 - 183*2 = - 24*0
Contraction from calculated value A = 207*2 - 181*7 = - 25*5
* The compound with this density was called by Schiff terpene, but is probably
pinene.
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT . 63
The theoretical contraction is on the basis of the plane
formula:—
One four-membered ring - 8*5
One six-membered ring - 15*0
There is also a difference between the contraction deduced
from the data and that indicated by the summation method, as
for camphene, but it is not so great.
Fenchene C 10
CH a -CH- C
(CH,),<i
ch,—-!;h—<
Two five-membered rings A = — 354*0
The difference in this case is (26*4 - 24) = - 2*4
The result may be due to the grouping
-C-™.
- CH
which in ft amylene shows a contraction of - 2*4. The idea is
that the association of methyl groups and | = | gives rise to special
contractions.
It is probable that the grouping
C
which is found centrally, is responsible for a contraction of
about - 1 ’0, as in tetramethyl methane
CH 3
ch 8 —i—CH«
It is seen that camphor, borneol, and the parent hydrocarbons
possess contractions which are very much greater than for single
six-membered rings. In fact, bridged rings of some kind are
denoted, since, as we have seen, the types of double rings dis¬
tinguished by cross-linking are not subject to very much larger
contractions than for single six-membered rings.
Digitized by v^ooQle
6 4
LIQUID CHEMICAL COMPOUNDS .
(d) The Sesquiterpenes (C 6 Hg) 3 .
We have noticed the following stages in the polymerization
of isoprene C 6 H 8 .
(i) Two molecules forming the menthan terpenes with a
single ring.
(ii) A bridging of the ring.
A further stage is indicated by the union of three molecules
of isoprene to form the sesquiterpenes.
Although the compounds C 16 H 24 are all classed as sesquiter¬
penes, it by no means follows that they are all alike in structure.
Indeed it is probable that many types of structure exist.
In any case the value of C = 0 46 may still be employed for
the ordinary type of sesquiterpene.
The sesquiterpenes C 15 H 24 .
Clovene b.p. 262 d 16 0*932
Cedrene „ 261 „ 0*932
Ledene „ 264 „ 0*935
For clovene Vm'= 265-6, and a similar value for the others.
By the summation method,
[C 16 H 24 ] = 310*8 (84 X 3 * 7 )
Vm = 265*6
A = ~ 45*2 (3 x - 15*0).
Thus the condensation of three molecules of isoprene and
the presence of three six-membered rings would account for the
molecular volumes obtained.
It is obvious that if the complex molecule consist of three
molecules of isoprene, then it must include a bridged ring, that
is, a ring which contains a central group linked to two of the
peripheral carbon atoms. Otherwise only two six-membered
rings would be possible. For our present purpose it is im¬
material which of the camphane terpenes forms the basis of the
sesquiterpene molecule. We shall assume that it is camphene.
Under these circumstances the formula would be
CH a CH S
I I
CH a -CH-CH CH S -C CH*-CH-CH—CH 2 —C
CH 3 l—CH 8 -> CH3J—CH3
CH*-C-CH CH 3 -CH CH 2 -C-CH—CH 2 — OH
camphene. isoprene. sesquiterpene.
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT
65
The above molecule should possess a volume of 268 0, which
approximately agrees with the number calculated. If the value
- 2-4, just shown to be necessary, be deducted, we obtain the
calculated value
2»V a 268*0 - 2*4 = 265*6
By formula
Vm = 265*6
Clovene terminates in an olefin linking, and this is capable
of taking on another molecule of isoprene.
Semmler in his Atherische Ole classes the following as
tri-cyclic:—
Aromadendrene
b.p.
260-5
o*g 249 1#
Clovene
261-2
o* 93 °i 8
Cedrene
261-2
o* 9359 i 5
Ledene
264
o *9349
Vetiden
255
0*9332
and doubtless there are others.
Cadinene is described as
/ \ / CH \ / \
CHg CH 3 CH a CHg
Digitized by v^ooQle
66
LIQUID CHEMICAL COMPOUNDS.
Such a compound as the last would have a volume of
#
Carvene C 10 H 16 190*5
Isoprene C 5 H 8 103*9
less for six-membered ring
294*4
~i5‘o
Volume of cadinene 279*4
(c) Diterpenes (C 5 H 8 ) 4 .
Among the diterpenes we find
Diterpene b.p. 320° d 16 0*9535 d* 0*9414
Colophene 318-320° d 1B 0*940
Still utilizing the value C = 0*46.
d b.p. 0*7714 Vm 352*4
we obtain the following results.
By summation,
[CioHaJ = 414*4 (112 x 3*7)
V m = 352*4
A = -62*0 (4 x - 15*5).
A better method would be
^10^90 4*4*4
less for ring - 6o*o
less for CH 8 |
354*4
-2*4
%Na 352*0
Vm 352*4 (by form.).
A formula which would answer to the above description
is the following:—
ch 8
-CH—CH,—i CH,= CH
CH
CH-
CH-*
H 9
-CH—CH a —CH CH a =C
£
ch 8
CH,-C-CH—CH,— A —CH,-CH
a'
CH,-
CH,—
,-L.
CH,
Ah,
-CH—CH,—CH—CH,—C
Diterpene.
^H,
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT.
67
A molecule of this character should possess a volume of
about 352*0, and, as it terminates in an olefin link, the poly¬
merization process could continue until compounds of the com¬
plexity of india-rubber might result.
The Effect of Cross Linking. The Double Bond.
It has been stated that compounds which show the cross-
linking, as distinct from those which possess bridged rings, are
marked by contractions very little different from those which are
characteristic of single rings.
The following are examples of this type of structure with the
contractions appropriate to them :—
// c \
0 / o
c
O G
(I)
(2)
* - {:£)
- = -18*0
A a - 15*0
.* = {:£}
(
-
C
c
/
\
/ \
/l\
c
C
c c
c c
A
1
c
A A
if i
\
./
\/
N/
( 4 )
(5)
(6)
*-{:!?
1 = -17-0
A =* -15*0
_//°V
N,/
/ c \
0 c
It is also seen that ring (3) does not differ from ring (6) as
regards contraction.
The cross linking does not produce any marked effect
additional to that produced by a single six-membered ring,
provided that it does not include a C or other atom centrally
situated .
If it does include such an atom or groups a very marked in¬
fluence is exerted on the volume in addition to that produced
by the former type of ring, due to an attractive effect of the
central carbon atom on the peripheral atoms, ar\d th$ qorrespond-
ing reacting effect on the central atoms.
5 *
Digitized by v^ooQle
68
LIQUID CHEMICAL COMPOUNDS,
In the first case , that of simple cross linking, although it
involves the existence of two rings, can scarcely be distinguished
from a single ring of the same outline as the first.
This is unfortunate from the point of view of chemical di¬
agnosis; but if we ascertain beforehand the number of olefin
linkings present, by the method of molecular refractions, no
difficulty will be experienced in giving the compound its true
structure.
The boiling-point is also distinctly lower than for a single
ring terpene, 150° or so as against 170° - I76°C.
The Olefin Link.
It probably will be a matter for remark that the olefin link
does not occasion any apparent contraction at the boiling-point.
If we consider the curve showing the contractions for a series of
ring compounds—the polymethylenes, for example, we shall
find that on extrapolation from Diagram (2) p. 48, the olefin link
would correspond to a considerable contraction.
There is, however, found to be no contraction. This result
calls for an explanation.
v
c
/ c
c
Nc
A + o
or - 3*5 by extrapolation.
A - 5*0
A - 8*5.
If we apply the same principle to the double bond as was
brought out by cross-linked rings, we find an explanation of the
zero value.
The principle involved is as follows: (a) In simple cross
linking the changed direction of the link probably does not
by itself involve any change in the shape or structure of
the molecule, and thus there is no great change in magnitude
of the contraction. It is only when an atom or group are in¬
cluded, that a considerable modification takes place in the shape.
Under these circumstances the magnitude of the contraction is
greatly changed.
( b ) In the olefin type of linking, or for that matter the acetylene
Digitized by
Google
HYDROCARBONS AT THE BOILING-POINT 69
type, no change in the structure is discernible from that char¬
acteristic of the saturated atom for the reason stated.
- ch 2 - ch 2 - - ch = ch - -c = c -
When, however,
CH S - CH = CH, changes to ^CH,
CH/
a modification in the constitution has occurred, which is at once
noticed in the volume.
In the ring system
I H
it is evident that structure I (J?) does not occasion any marked
change in the shape or volume of the molecule, because the
affinity is directed to the penultimate carbon atom, instead of to
the neighbouring one. It is simply a case of change in the
direction of the affinities. The shape and volume of the molecule
are chiefly determined by the single linkings of the six-membered ring.
In II, there is included a carbon or alkyl group, and this
exercises a modifying influence on the volumes of the peripheral
carbon atoms, and so on the molecule.
In compounds with an olefin link the atoms are not more
favourably situated to act on each other than in the saturated
chain. No change in the volume occurs, that is, there is no
special action due to unsaturation, at least under the circum¬
stances.
In the trimethylene ring , as compared with an open-chain
compound with an olefin link, the atoms are more favourably
situated for interaction, owing to the greater concentration of
matter, for the change in the direction of one of the carbon
valencies, involves a change in the position of one of the methy¬
lene groups, relative to the others.
Thus owing to the distortion of the original straight-chain
molecule, there is a contraction in volume.
Digitized by v^ooQle
7 °
LIQUID CHEMICAL COMPOUNDS.
This occurs
when CH 8 ~CH 2 —CH 2 —CHa—CH 3
changes to
CH a \
CH S /
CHj—CH a —CHj
when
CH S —CH a —CH 2 —CH a —CH a —CH 3
changes to
or when CH 3 —CH = CH a CH aX
| >CH 2 .
changes to CH a /
In all these cases the affinity forces are more effective than in
linear arrangement because of the greater concentration of matter.
It is also for this reason that the structure which has been
named partial ring structure involves very noticeable contrac¬
tions, although not so large as for completed rings. The curva¬
ture of the hydrocarbon chain results in diminished volumes,
owing to the greater concentration of matter, as compared with
what we call straight-chain compounds.
These results are thus just what we should expect, if we
assume that the carbon atoms or hydrocarbon elements of the
molecule exert an action of affinity on each other, not only in
the direction of the linkings but in other directions as well. The
greater relative efficiency of the attractive forces makes it impos¬
sible for the expanding heat forces to be so effective, and so there
are contractions.
It would appear that the above facts can be included in a
single generalization which may be stated in the following
terms:—
When any molecular change involves a change in its configur¬
ation! that is when one or more atoms are displaced relatively to
the others , then there is a distinct effect on the molecular volume .
When , however, the change includes simple displacement or a change
in the direction of a linkage only , no marked effects on the molecular
volume occur .
Conversely , modifications of the molecular volumes of compounds,
other than those due to simple changes in composition , may be
referred to changes in the shape of the molecule .
CH a
/ \
ch 2
ch 2
(Ihj
Jhj
\h/
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT
7 i
This generalization apparently covers all, or nearly all of the
constitutive effects which have been noted in connexion with
carbon compounds and is doubtless of great importance.
Ring Systems . 30
Ring systems are divisible into the following classes:—
(a) Simple ring systems. Example, Benzene.
(&) Multiple ring systems.
(i) Condensed—naphthalene.
(ii) Separated rings—diphenyl.
If n be the number of carbon atoms in the compound and
N the number of atoms or groups in the ring, we are able to
find the nature of the molecule by the amount of the contrac¬
tion per atom given in the following table:—
TABLE XXVIII.— Contractions, etc., for Different Simple Rings.
No. of atoms
in ring.
N
^ . Contraction
Contraction. ^
Three-membered
3
- 6-0 “2’0
Four-membered .
4
- 8*5 - 2*12
Five-membered .
5
-12*0 - 2*40
Six-membered .
a
- i5’o - 250
Seven-membered
7
- 20*8 - 2*97
Eight-membered
8
-26*0 -3*25
The last column is especially useful in distinguishing condensed
from separated rings.
In these, if the number of C atoms or others like O, S, N be
n and the number of groups, as measured by the contractions be
N, then N > = or < than n.
{a) Condensed Rings.
Example, naphthalene c 10 h 8 .
In this case N - n = a positive number.
The contraction is - 30*0. Dividing this by an average value
of Tj = 2*50, we find that the number of groups by calculation or N
30*0
2*5
12.
N - n = 12 - 10 = 2, that is N ^ n by 2.
Digitized by v^ooQle
72
LIQUID CHEMICAL COMPOUNDS.
Since the number of carbon atoms in C^Hg or n is io*o, then we
conclude that naphthalene contains a double-condensed ring system^
and that there are two atoms common to both rings.
CH CH
CH CH
✓ \/
CH C*
1 H
CH C*
\ /\
CH CH
\h
CH
Naphthalene.
s \/ x
CH C* CH
Ah c** Ah
x /\ s
c* c*
l',H a -llH,
Acenaphtene.
In acenaphtene C 12 H 10 (condensed ring) the contraction is
- 45 0, and a similar calculation can be made.
— = 18 or N
2*5
N-n=i8-i2 = 6
There are at least three rings in the compound.
If we observe the diagram, we shall see that three carbon
atoms are marked by * and one by **. This is to indicate that
in the one case three carbon atoms are common to two rings,
and in the other one carbon atom is common to three rings.
This is somewhat more than a formal statement, because it
indicates that the three common C atoms are subject to double
the compression that the other or peripheral carbon atoms are,
and in the other case the single carbon atom has thrice the
compression.
There is a fundamental principle at the back of these con¬
tractions, which is, that the mutual attractive influences exist,
which result in diminished volumes. The result is, therefore,
that if the contraction is contributed to by the action of one
carbon atom on all the others, it, in a similar manner, is acted
upon by the other atoms, and so the contraction is doubly
affected.
It is, however, to be observed that the C atom of one ring
does not influence the carbon atom of another ring unless it
forms part of this ring.
The presence of side chains, of course, complicates matters,
but in this instance
n > N
so that the formula for the compound can be fairly well made
out.
Digitized by v^ooQle
HYDROCARBONS AT THE BOILING-POINT.
73
(b) Separated Rings.
Example, diphenyl C 6 H 5 - C 6 H 6
In this case N - ft = O
A calculation similar to the above will indicate the nature of
the ring system.
The existence of other constitutive effects besides that of ring
structure of course makes it more difficult to assign a formula.
Digitized by v^ooQle
CHAPTER III.
THE MOLECULAR VOLUMES OF ORGANIC COMPOUNDS
CONTAINING THE HALOGENS.
The Halogens.
KOPP calculated the volumes of chlorine, bromine, and iodine in
the combined state as :—
[Cl] - 22-8 [Br] - 27-8 [I] =* 37-5,
which he showed were not very different from their free values.
[Cl] = 227 [Br] - 26-8 [I] - [34-0]
(Kopp) (Thorpe). (Billet).
Those proposed in the present treatise are similar:—
[Cl] = 22-1 [Br] = 27-0 [I] = 37*0.
The halogens are all monovalent, and in this respect, as
perhaps in others, they resemble hydrogen.
Fluorine F.
Compounds of the type R - F have not been investigated.
Only one or two data for fluorine compounds exist altogether.
They are:—
CgH 0 F ioi*6 (Young)
UgHg 92'8
F = 8*8
As F s 53*8 (Thorpe)
As 27*8
3 F = 26*0
F = 8*7
In order to confirm, if possible, the above value, a few com¬
pounds, indiscriminately chosen from chemical literature, have
been studied by means of the formula.
Before calculating the values, it will be necessary to know
those values of C which are applicable to such compounds.
74
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING THE HALOGENS . 75
TABLE XXIX.
Substance.
d b.p.
c.
CH a : CH . CH,C 1 .
do
0*9610
d«
0*9002
0*468
CH,. C 0 C 1 ....
tt
1-1377
d„
I*0570
0*482
CH a Br. CH,Br .
tt
2*3132
dm
1-9312
0*450
Mean value of C
•
0*467
It should also be noticed that when atoms like F, Cl, Br and
I occupy the a and ft positions in the hydrocarbon chain, there
are contractions. These contractions have been shown to be due
to a curved configuration of the hydrocarbon chain (vide prox).
Thus:—
CH 2 -CH 2 -X
I N
»
;
CH 2 -CH 2 -y J
It will be shown that the above configuration is responsible
for a contraction, the magnitude of which is dependent on the
size of the partial or potential ring. A few such examples
are:—
TABLE XXX.
Substance.
Vtn.
2»V a .
A.
CH Br : CH Br .
91*4
93-0
- i*6
CH a I . CH a Cl .
101*3
103-3
-2*0
Mean value
.
-1*8
TABLE XXXI.— The Volume of Fluorine (by Calculation).
Substance.
b.p.
d /.
Vm.
Sum. of
at Vols.
-i*8.
F.
CH F a . CH a I .
89*5
2*2412,,
94*1
65*9
2 x 9*1
CH F Br . CH 2 Br .
121-5
2 ‘ 2 6 33 io
103*0
95*9
Mean for F
.
8*4
Data due to F. Swartz. 31
Digitized by v^ooQle
76
LIQUID CHEMICAL COMPOUNDS.
If we consider the probable errors incident to this indirect
method of calculation, we must conclude that the result is in
good agreement with the former values.
The value already obtained for fluorine, viz.:—
[F] - 87
is thus the correct one.
Chlorine Cl.
In considering homologous series such as those of the alkyl
halogen compounds, it should be borne in mind that the atomic
volumes, calculated by the usual method of differences, are liable
to be affected by influences which are of a constitutive nature,
and which really act on all the atoms of a molecule. It follows
that the results obtained by this means are likely to be different
from the true values.
The errors to which this method of calculation is liable, are
considerably diminished, if we take the average of a large number
of results, but this really amounts to ignoring differences below a
certain magnitude.
There is an alternative method which is not based upon com¬
parisons at all, but which enables us to calculate the atomic volumes
from a consideration of individual compounds. This method
depends upon a principle found to hold in the hydrocarbon
series, and which is, that the relative volumes of the atoms are
maintained in any series of similarly constituted compounds.
Thus the relation [C] = 4[H] is generally true. The principle
in question is that of constant atomic volume ratios .
It is possible that the ratios between the atomic volumes may
not be quite the same as those assumed, but this error would not
appreciably alter the form of the curves obtained. These conse¬
quently indicate the nature of the changes in the volumes of the
atoms from compound to compound in a series.
Hydrochloric acid HC 1 .
By D. Bertholet’s formula, V = -^r = ** * ^ * .
J D P k (2T k -T)
Tk = 325 0 C P K = 86 atmos. T b.p. 238° C.
Vm = 33 ’i.
The volumes of the following compounds are considerably in
excess of the normal.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING THE HALOGENS . 77
TABLE XXXII.
Substance.
V m .
2 »Va.
A.
HC 1 .
33 *i
25*8
+ 7*3
CHg . Cl .
50*4
48*0
+ 2*4
C a H 6 Cl . .
717
70*1
+ 1*6
Such abnormally large volumes are characteristic of the
initial members of most series. In these cases they may be due
to the influence of the hydrocarbon and hydrogen elements, since
the volumes of methane and ethane are considerably in excess of
the volumes which we should expect them to have, from a com¬
parison with succeeding members of the paraffin series.
In any case, it is probable that the volatile hydrogen is
responsible for these augmentations. It will be borne in mind
that the free and combined values of chlorine are similar.
Free value* [Cl] = 227. Combined value [Cl] = 22*1.
TABLE XXXIII.— The Alkyl Chlorides.
Substance.
Vm.
Vols. of
Hydrocarbon
Groups.
Cl.
C 8 H 7 C 1 .
91*7
70
217
C^Cl .
ii 3*5
92*3
21*2
c 6 H n ci. .
135*3
114*1
21*2
Mean value [Cl]
= 21*4
It is probable that chlorine, like carbon, is an integral number
of times larger than hydrogen. This is seen from the following—
C 8 H 7 C 1 917 C 4 H 9 C 1 113*5
C 8 H 6 C 1 84*4 CgHnCI 135*3
2[H] = 7*3 [CHJ = 21*8 = 3 X 7*3 (approx.).
Since [Cl] = 21*4 = 3 x 7*2 (approx.)
It follows that [CHJ = [Cl] = 6[H].
This relation enables us to find the number of hydrogen
equivalents W in a series, and so the volume V/W of such an
Digitized by v^ooQle
78
LIQUID CHEMICAL COMPOUNDS,
equivalent. These values show how the volume of the atom
hydrogen varies in a particular series.
TABLE XXXIV.— The Volumes of V/W in the Alkyl Chlorides.
Substance.
w.
V w .
V/W.
HCl
7
33 ‘i
4728
C 3 H 6 C 1 . .
19
717
3*774
c 8 h 7 ci . .
25
917
3*668
C 4 H # C 1
31
ii 3*5
3*661
C 8 H u C 1 . .
37
135*3
3*657
The Unsymmetrical Polychlorides .
The successive substitution of chlorine for hydrogen in
methane (CH 4 ) or ethane (C 2 H 6 ) does not correspond to equiva¬
lent differences in the volumes or the boiling-points.
TABLE XXXV.
Substance.
b.p.
A b.p.
for Cl-H
Vm .
A for
Cl-H
CH 4 .
-164°
380
140
124
CH S C 1 ....
- 24 °
50*4
65-6
M *7
CHjCL, ....
+ 41*6°
651
19*6
19*4
CHC 1 8 ....
6l"2°
84*5
15*5
19*2
CC 1 4 .
767°
103-7
ch 8 .ch 8 .
- 93 °
567
105-5
15*0
ch 8 .ch 9 ci
4-12-5°
717
71*0
17-2
CHj.CHCJj .
83 - 5 °
88-9
iro
19*1
CH 8 .CC 1 8
74 - 5 °
108-0
We see that in both series there is a decrease in the boiling-
point, and an increase in the volume for every substitution of
chlorine for hydrogen.
In the following table which shows the values of chlorine in
the different compounds, we assume that the volumes of carbon
and hydrogen are similar to their values in the hydrocarbons.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING THE HALOGENS . 79
TABLE XXXVI. —The Polychlorides.
Substance.
Vm.
Volume of
Hydrocarbon
Group.
n Cl.
R-Cl ....
21*5 (average)
CHjCl, ....
65*1
22*1
2 x 21*5
CH Cl,
84*5
18*5
3 X 22*0
CC 1 4 .
IO37
14*8
4 x 22*2
CHj.CHCl a . . .
88*9
44*5
2 x 22*2
CH 8 .CC 1 8
io8*o
40*8
3 x 22*4
Since [Cl] = [CHJ
[Cl] = } x [CH S CH C 1 J = } x 88*9 = 22*2.
This value is similar to the one found in the table by another
method.
We conclude that the average volumes of the chlorine atoms in¬
crease as they accumulate in molecules of the above type .
As an alternative to the above, we may assume that two
chlorines possess volumes of Cl = 21*5 and two of volumes
Cl = 23 o. The above assumption is, however, the simplest.
Bromine Br.
The free value according to Thorpe is—
2[Br] = 2 x 2674.
In the monobromides we find—
Substance. V . 2 *V a (R). A.
CH 8 Br 557 260 297 + 2*6
This positive value represents the excess above that which
may be considered normal.
TABLE XXXVII.
Substance.
Vm.
R.
Br.
C 8 H 7 Br . .
9 T 3
70*0
27*3
C 4 H 9 Br .
118*6
92*3
26*3
C 8 H ai Br .
141*2
114*1
27*1
The average volume of bromine is about
[Br] = 27*0,
Digitized by v^ooQle
8o
LIQUID CHEMICAL COMPOUNDS,
The Poly bromides.
TABLE XXXVIII.
Substances.
V m .
Br.
R-Br .
26*9 (average)
CH 2 Br a .
77*6
27*7
CHBr 3 .
103*5
28*3
CBr 4
(Not studied)
(probably larger)
Mean value .
. 27*6
The observation that the substitution of a negative atom for
a positive one involves an increase in the atomic volume, is
repeated in the polybromides, provided these halogen atoms are
attached to the same carbon atom.
The values for bromine which have been shown, increase pro¬
gressively as the number of such atoms united to a single carbon
atom increases. The volume of bromine may, as already indi¬
cated, tend to a higher value, viz. 28 o or even 28*5.
The following compounds are noteworthy :—
CBrClg 108*4 CC1 8 8o*8 Br 27*5
POBrClj 107*38 POClg 101*37 - 22*0 = 79*37 Br 28*0
Cl
Br
1
Cl—C—Cl
1
H—C—Br
1
Cl
1
Br
[Cl] = 22*2
[Br] = 28*3
Cl—Hg—Cl
[Cl] = 22*1
Cl
Br—Hg—Br
[Br] - 28*5
Br
[Hg] = 18-9
1
ci—Si—Cl
1
|
Br—Si—Br
1
Cl
[Cl] = 22*2
Br
[Br] = 28*0
[Si] = 34-0
These numbers show that in an ordinary chloride or bromide,
the value for chlorine is 22 T, and that for bromine 28*0.
One remarkable exception is found, viz. P Br 3 .
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING THE HALOGENS. 81
Br
Br—P
>
P 27‘0
3Br 84*0
2nV a liro
Vm 108*3
Br
A = - 27
The volume of bromine is seen to be smaller than in most of
the other bromine compounds, in which it is 28*0.
If we suppose that
[Br] = [P] = 27*0, then
2«V« = [P] + 3[Br] = 27*0 x 4 = 108*0
The similarity between the combined volumes of phosphorus
and bromine is probably accidental. The important point to
notice is that the three bromine atoms assume similar volumes to
the value in R - Br, which has been shown to be 26*9. This
involves a failure of the usual law of increase with the accumula¬
tion of halogen atoms about a single carbon atom for some reason
or other.
Iodine I.
The free volume is according to Billet—
2[I] = 2 x 33*5.
Methyl iodide.
CH 3 I V w = 63*9
CH S 26*0
I 37*o
2«V a 63*0
Vm 63*9
A = + 0*9
This result is in accord with those previously obtained for
the other series, and is due to the presence of the methyl group
as shown in the following table:—
TABLE XXXIX.
Substance.
Vm.
2 *V a .
A.
CH 8 - H .
38*5
29*6
+ 8*9
CH 8 -C 1 .
507
47*6
3 ’ 1
CH 8 - Br .
557
53 *°
27
ch 8 -i
63-9
63*0
0-9
6
Digitized by v^ooQle
82
LIQUID CHEMICAL COMPOUNDS,
These plus differences are found to diminish with the apparent
size of the atoms, etc. They also diminish with the number of
CH 2 groups, and thus they nearly disappear at about the third
member of each series.
The monoiodides are:—
TABLE XL.
Compound.
V*,.
Volume of
the Hydrocarbon
Radicle.
I.
V/W.
CjHjI . . .
85*8
48*0
37-8
3730
CjHjI . . .
106*8
70*0
3 6*8
3-683
C.HjI. . .
128*2
92*0
36*2
3«3
C 6 H u I . .
151*4
114*1
37*3
3693
c«h 6 i . . .
130*5
92*8
377
Mean value [I] .
37 *o
The average value of iodine is—
[I] = io[H] - 37*0,
which is a relation enabling us to find the variation in the atomic
volumes throughout the series.
TABLE XLI.— Values of V/W in the Alkyl Iodide Series.
Substance.
w.
Vm.
V/W.
CH,I . .
17
64*1
3770
CjH s I . .
23
85*8
3730
c,h 7 i . .
29
106*8
3-683
c 4 h,i . .
35
128*2
3-663
C,H U I . .
4 i
151*4
3’693
Several higher compounds have been studied by Dobriner,
but the curve is not very regular.
In this series, we note similar characteristics to those for the
paraffins.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING THE HALOGENS . 83
(a) A gradually diminishing series of values to—
(b) a minimum at the fourth or fifth member;
(c) an increasing series of values after this minimum.
A similar result is shown in Table XL, col. 4, so that the
nature of the curve indicated by the values of V/W is the true
one.
In dealing with the halogens, we may include an inorganic
series examined by Prideaux.
HgCl a 63-3
Hg 189
cl* 44*4
2*V a =63*3
HgBr* 75*1
Hg 18*9
Br a 56*0
2*V a =74*9
Hgl a 928
Hg 18*9
I 74*0
5«Va = 92*9
The values of Cl, Br, and I are—
[Cl] = 22*2 [Br] = 28*0 [I] = 37 # o.
in an inorganic series.
These are similar to those already shown for organic com¬
pounds.
The average volumes of the halogen atoms are thus—
V« [F] = 87 At. wt. 19*0
[Cl] = 22-o „ 35*5
[Br] = 28*0 „ 8o*o
[I] = 37*0 „ 127*0
In this family the atomic volumes increase with the atomic
weights, but not proportionally.
Constitutive Effects among the Halogen Compounds.
The Combined Influence of a Halogen Atom and Iso
Group.
The effect of the branching of the hydrocarbon chain when
halogen atoms are present, has now to be studied.
Among the hydrocarbons (paraffins), a branching of the chain
is, so far as we can see, uniformly accompanied by a contraction.
If, however, one of the hydrogen atoms be substituted by a
halogen atom, a different result ensues. The following table
shows this.
6 *
Digitized by v^ooQle
8 4
LIQUID CHEMICAL COMPOUNDS,
TABLE XLII.
Chlorides,
Vm,
A.
Vm.
Iso, etc. Compounds.
CH S . CH a . CH a . Cl .
CH S . CH a . CH 2 . CH a . Cl
CH S . CH 2 . CH 2 . CH a . CH 2 C 1 .
917
1127
135*3
+ 2*6
+ i*8
-0*5
94*3
114*5
134*8
(CH 8 ) 2 CHC 1 .
(CH 8 ) 2 . CH . CH 2 C 1 .
(CH s ) 2 .CH.CH 2 .CH 2 .C 1 .
Bromides,
CHj . CH 2 . CH a . Br .
CH S . CH 2 . CH a . CH 2 . Br
CHj . CH 2 . CH 2 . CH a . CH 2 Br
97*3
118*6
141*2
+ 2*2
+ 1*0
-0*4
99*5
ng*6
138*8
(CH 8 ) 2 . CHBr.
(CH 8 ) 2 . CH . CH a Br.
(CH 8 ) 2 . CH . CK a . CH a . Br.
Iodides,
CH S . CH a . CH 2 . I .
CH 8 . CH a . CH 2 . CH a . I .
CH S . CH 2 . CH 2 . CH* . CH 2 I .
107*1
128*2
151*4
+ i *5
+ 0*2
- 0*3
108 *6
128*4
151*1
(CH 8 ) 2 CHI.
(CHI, CH . CH a . I.
(CH s ) a . CH . CH 2 . CH a . I.
We see that in the case of the propyl and butyl halogen
compounds, the iso or branched chain derivatives are larger than
those with straight chains. Exceptions to this are found among
the amyl compounds, which show contractions approximately
equal to that for the iso-group.
The positive differences thus seem to be connected with an
approach of the halogen atoms to the iso-group.
Amyl chloride (iso)
Butyl chloride (iso)
Propyl chloride (iso)
CH,
CH—CH,—CH,—Cl
Ah,
or CH,—CH—CH,—CH,—Cl
i
H,
CH,—CH—CH,—Cl
Ah,
CH, . CH . Cl
Ah,
A - 0*5
A + I‘8
A + 2-6
The inference is, that the approximation of a halogen atom
to a methyl group causes an expansion which diminishes as the
two groups become separated, until, when sufficiently removed,
the contraction for the iso group alone remains.
We have thus to admit the possibility of the interaction of groups
not directly connected by valency linkings .
Whatever be the exact cause of the expansions, it would seem
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING THE HALOGENS. 85
that they take effect on the halogen atoms, that is, their volumes
increase by the values J. Thus in the compounds
CH 8 - CHX - CH 3 , [Cl] - 21-5 + 2-6 = 24-1, [Br] = 27-0 + 2*2 = 29*2, [I] = 38*5
This fact is proved by an examination of the compounds
of formula
CH 3 - CXj - CH 8 (CH a ) x d 16 = ri<>58, C = 0*50
from ch 8 .ci.chh
Vm 114*8
2CH 8 52*0
C 14*8
66*8
Vm 114*8
2[C1] = 48*0
[Cl] = 24*0
(CH 8 ) 2 CBr 2 , do 1*8149, b.p. 114*0, C = 0*463
from CH 3 . CH 2 Br. CH 2 Br
V w 126*4
Vm 126*4
(CH 8 ) 2 C 66*8
2[Br] = 59*6
[Br] = 29*8
These facts are of the greatest importance. They show that
an atom may have more than one value according to its position,
in a hydrocarbon chain.
Calculation shows that the above expansions are much aug¬
mented in compounds containing the tertiary group—
ch 3 —ch 9 —
<!:h.
a, /?, and y Halogen Compounds—Partial Rings.
The di- and tri-substituted paraffins in which the halogen
atoms are attached to a single carbon atom
e.g. CH 3 X, CH 2 X 9 , CHX 3 , CX 4
CH s . CH a X, CH 3 . CHX,, CH 8 . CXg
have been studied.
In all of these a compounds the additive rule is approxi¬
mately realized. Other compounds, however, occur, in which
the halogen atoms are attached to different carbon atoms, e.g.
X - (CH 2 ) w - Y.
This mode of distribution is found to be accompanied by
very considerable modifications in volume.
Digitized by
Google
86
LIQUID CHEMICAL COMPOUNDS.
TABLE XLIII.— Polyhalogen Compounds.
a, j 3 , and y derivatives.
a Compounds.
V m or 2 n V a .
A.
Vm.
a, £, etc., Compounds.
CH, . CHCl,
CH, . CH, . CHCl, .
CH, . CHBr, .
CH,. CHC1I .
88*96
2»V a iii’o
-3*6
- 3* 1
85-3
107-9
CH,C1. CH,C1.
CH, . CHC1. CH,C1.
100*2
-3*i
97*06
CH,Br . CH,Br.
103*6
-2*6
101*0
CH,C1 . CH,I.
CH, . CH, . CHBr, .
»»
122*4
-3*2
119*2
CH, . CHBr . CH,Br.
CH,. CC1, . .
»»
107*0
-3*3
103*7
CH,C1. CHCl,.
C,H,C1 4 .
125*4
-60
119*4
CHC1, . CHCl,.
C,HC1, . . .
1438
-5*6
138*2
CHCl, . CCl,.
C,C1 4 ....
118*0
-3*2
114*8
CC1, : CCl,.
CHBr : CHBr.
C,H,Br, . .
93*<>
-1*6
9i*4
CH, . CH, . CHCl, .
CH, . CH, . CHBr, .
111*0
122*4
-5*0
-4*3
H H
s»&
H 6
CH,C1 .°CH,. CH,C1.
CHjBr . CH, . CH,Br.
It is evident from the above that
(a) A halogen atom in the (/*) position, counting from the
end of the carbon chain, occasions a contraction of 3 units, but
only if one or two halogen atoms are found in the (a) position.
Halogen atoms in the (a) position alone or in the Q8) position
alone do not cause contractions (vide ante).
( b ) Given the requisite number in the (a) position, the con¬
traction depends upon the number in the (£) position. This is
seen from the fact that in tetrachlor and pentachlor ethane, the
contractions are double, or amount to 6 0 units.
Conclusion : The only explanation which we can give to the
above facts is, that there is some sort of interaction between the
two chlorine atoms, which is due to their being contiguous.
This necessitates such an arrangement of the atoms as to allow
of this, and the contraction depends upon the number of groups
in the partial ring.
This is shown by means of the plane formulae as follows:—
Ethidene chloride.
Cl
CH
4 ,
H
Ethylene chloride.
CH 2 -Cl x.
CbU
-Cl
A = - 3 * 6 .
\
1
1
/
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING THE HALOGENS . 87
Propidene chloride.
Cl
Cl
Propidene chloride.
ch 2 - a \
I )
CH -Cl' 7
I
CH3
A = - 3 * 1 *
Trichlor ethane.
Cl-CH-CK
iH.,
-Cl*'
Trichlor propane (chlorhydrin).
CII 2 - 01 \
CH -
cl 2 -
-Cl
*ci'
A = - 3*1.
A = - 6*0 probably.
See Chapter II, p. 52.
Tetrachlor ethane. Pentachlor ethane.
/C\ -CH- CB n ,,C1-CH- C\\
\
a
CH-
gi
'Cl-
cr
A = — 6*0.
Cl
A = - 5 * 6
(^r) If we study the trimethylene derivatives, we find that
the contractions are largely increased. Since the removal of a
halogen atom to the 7 position would separate it from the one
in the a position, and the large contractions indicate a correspond¬
ing interaction, we must suppose that the configuration of the
hydrocarbon chain is modified in accordance with these results.
This involves an arrangement of the carbon atoms on a curve,
or the formation of a partial ring.
A =-5.0 ^ = -4.3
Although the result is doubtless due to an approximation of
the halogen atoms, it seems clear that all the atoms share in the
above modification. Probably 8 and e compounds would show
still larger contractions,
Digitized by v^ooQle
88
LIQUID CHEMICAL COMPOUNDS,
(d) The presence of an ethenoid linkage influences the amount
of the contraction.
It has been shown that in compounds CH 3 . CHX . CH 3 and
CH S . CX 2 . CH 3 , as compared with CH 3 . CH 2 . CH 2 X and
CH 3 . CH 2 . CHX 2 , X being halogen atoms, etc., the volumes of
these atoms are larger in the /8 position than in the a . It follows
that if there is a contraction when one appears in the a and also
one in the /8 or 7 positions, it must be due to some relation
between these contiguous atoms.
Since in the ay compounds, the contractions are greater than
in the that is, when they are farther removed, the large con¬
tractions can only be attributed to a change in the configuration
of the hydrocarbon chain brought about by the interaction of the
terminal atoms.
The Volumes of the Unsaturated Polyhalogen Compounds.
There are evidences that on passing from the saturated to
the unsaturated polymethylenes, a constitutive change takes
place, which is connected with the combined influence of the
olefin link and the halogen atoms. Only one or two data, how¬
ever, exist, so that it is scarcely possible to obtain that extent of
view which leads us to certain conclusions. It will be thus
necessary to resort to calculation to supply the deficiency to
some extent.
Chlorides.
Ethyl chloride Vm.
CH 8 — CHjCI 717
Vinyl chloride Vm.
CH a = CHC 1 —
Ethylidene chloride.
CH S —CHCl a 88*9 (obs.)
C = 0*46 89*0 (by form.)
A = 8*9 for H a
Unsymm. dichlorethylene.
CH a = CClj, 79*9 (obs.)
8o*i (by form.)
Ethylene chloride. Symm. dichlorethylene.
CH a Cl—CHC 1 85-3 (obs.) CHC 1 = CHC 1 —
C = 0*46 85’6 (by form.)
Acetylene tetrachloride.
CHClj—CHC 1 2 —
120*6 (by form.)
A = 5*6 for Hj,
Tetrachlorethylene.
CC 1 2 = CC 1 2 114*8 (obs.)
115*0 (by form.)
Propylene chloride.
CH 3 —CH 2 —CH^Cl 91*7
Allyl chloride.
CH 2 =CH—CH 2 C 1 85*0
A = 6*7 for H 2
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING THE HALOGENS . 89
Ethyl bromide V«.
CH,—CH a Br 77 # i (by obs.)
nsymm. dibromethylene.
CBr, —
Bromides .
Vinyl bromide V m .
CH a = CHBr 71-3
: 5*8 for H a
Ethyledene bromide. Uns
CH,—CHBr, — CH, = CBr,
98*8 (by form.) 93*0 (by form.)
A = 5*8 for H a
Ethylene bromide. Symm. dibromethylene.
CH a Br—CH,Br 97-1 (obs.) CHBr = CHBr 91*4 (obs.)
97*1 (by form.) 91*3 (by form.)
A = 5*8 for H a
Tribromethane. Tribromethylene.
— CHBr = CBr a —
119*3 (by form.) 113*6 (by form.)
A = 5*7 for H a
Propylene bromide. Allylene bromide.
CH a —CH a —CH a Br 97*3 (obs.) CH, =» CH—CH a Br 90*8 (obs.)
A = 6*5
In the above table, we have a summary of the results for the
unsaturated polyhalides. We find in the derivatives of ethane a
difference of 5 7 for H 2 .
normal A for H a 7*4
actual A for H a 5*8
change in value 1*6
From this we conclude that there is an increase of v 6 in the
volume of the unsaturated compound, presumably owing to the
close association of the halogen atom or atoms with the olefin
links |= |
Acetylene tetrachloride.
CCl a - CCl a
Cl—C—Cl
II
Cl-C—Cl
2 [C] = 29*6
4 [Cl] = 88*8
118*4
for part, ring 5*2
113*2
for unsat. +1*6
2»Va 114*8
Vm 114*8
Vinyl bromide. Unsymm. dibromethylene.
CH, = CHBr CH a = CBr,
CH,—CH a Br
77*1
Br CH,—CH a Br
98*8
less 2H
7*4
CH, = C less H,
7*4
for unsat.
69*7
for unsat.
91*4
+1*6
+ i*6
2 nV a
7 i '3
93*0
Vm
7 i # 3
v w
93-0
Digitized by
Google
9 °
LIQUID CHEMICAL COMPOUNDS,
Symm. dibromethylene. Tribromethylene.
CHBr = CHBr
CH—Br
II 2CH 37-0
CH—Br 2Br 56*0
CHBr=CBr a
Br—C—Br 2C
II H
CH—Br 3Br
VO t>*
for part, ring
93 *o
- 3-2
for part, ring
117-3
- 3‘2
for unsat.
89-8
+ 1*6
for unsat.
114*1
+ r6
2 »Va
Vtn
91*4
9**4
2»Va
Vm
115*6
113*6
or better
Tribromethane
CH a Br—CHBr a
less 2H
119*3
- 7*4
for unsat.
1 ii*9
+ i*6
2»Va =
V =
H 3*5
113*6
The following show scarcely any effect for unsaturation :—
Ally! chloride.
CH a = CH—CH 2 C 1
C,H b 62-9
Cl 22*1
2rtV a = 85*0
Vm = 85*0
Allyl bromide.
CH 2 = CH—CH s Br
C3H5 62*9
Br 28*0
2«Va = 90*9
Vw = 90*8
From the above, we conclude, that, in the unsaturated group
—CH=CX—
where X is a negative atom like Cl, Br, etc., there is an ex¬
pansion of + 1 *6. Since this expansion remains the same what¬
ever be the number of halogen atoms present, we are probably
right in connecting it with the presence of the olefin linkage | = |.
In allyl chloride, for example, such a feature is not noticed.
This shows that if the halogen atom be sufficiently separated
from the olefin link no expansion occurs.
In considering this explanation, we are led to find in it a
reason for a somewhat similar phenomenon connected with the
—C = O group.
In the aldehydes,
H
/
R—C
%
O
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING THE HALOGENS . 91
we shall find that the volume of O" is 7*4 (Chap. IV.), but in
certain other compounds, which involve the substitution of
hydrogen by chlorine, hydroxyl, and other negative groups, we
find a considerable
Thus in—
expansion.
Cl
OH
OR'
/
/
/
R—C
R—C
R—C
\
\
O
O
O
Acid chloride.
Acid.
Ethereal salt.
the volume of O" is 11 1, so that there is an expansion of + 37
as compared with O" in the aldehydes. Again we notice that
similarly there is unsaturation in the C = O group, as in the
—C = C— group.
We may thus suppose that these expansions are due to a
similar cause in the two cases. It is, moreover, not improbable
that certain well marked variations in the atomic volumes of
certain elements, which like oxygen show two or more values,
may also be due to the co-existence of unsaturation and negative
groups.
It must be observed that unsymmetrical dichlorethylene
CH a - CCla
is exceptional, in that there is a contraction instead of an
expansion* We do not propose at present to do more than
record the fact, as it is evident that we need much more
varied and numerous data before we can deal adequately
with the subject or come to certain conclusions. It is pos¬
sible that a repetition of the determination of the volume of
unsymmetrical dichlor ethylene and the determination of the
volume of symmetrical dichlor ethylene might show different
results.
General Remarks on Saturated and Unsaturated Poly halogen
Compounds .
The following represents a comparison of the contractions
in saturated and unsaturated derivatives.
* This seems to be due to the fact that the CH S group in CH S . CHC1 2 is
unusually large.
Digitized by v^ooQle
92
LIQUID CHEMICAL COMPOUNDS.
CH 2 -CK
I
c
-CH-
1
-Cl
• /
ch 2 Cl'
\
\
\pi_
1
_ _ r* 11
1
vl —
-
A = - 3-6
A = — 6.0
A = - 3*6.
A =
- 6’o.
CH—Cl
Cl—c—Cl
Cl—c—Cl
II
II
II
CH—Cl
CH a
0
1
0
>
II
l
ui
A = - 1*3.
A = - 2*6.
In regard to (2), we must also note the possible existence of
geometrical isomers, so that we might have
CH—Cl
II and
Cl—CH
Fumaroid form.
CH—Cl
II
CH—Cl
Malenoid form.
Precisely how these differ as regards volume is not quite
certain at present, but there is evidently here a rich field for
future research.
Applications of the above principles to a study of more com¬
plex compounds are indicated by the following.
We have noticed the following effects.
CH 8
£h—1
1
ch 8
C,H 7
70*0
ch 8
| 8
c 3 h 7
70*0
Cl
Cl
22*1
CH-Br
1
Br
28*0
Wm
92 # I
94*6
ch 8
2 nV a
v,»
98- 0
99 - 5
A
+ 25
A =
+ i *5
Thorpe 82 has given a volume of V w = 142*6 to the following
compound.
Isobutylene bromide.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING THE HALOGENS. 93
C 4 H 8 4 X 22*1
88*4
or
2Br
56*0
2CH3
52*0
. .-
C
14*8
144*4
CH 2
22*1
less iso group
-0*5
Br'
28*Of
Br"
29*2f
143 *9
.. ■
Br atom and iso group
+ 2*0*
146*1
- 3 *i
145*9
■ ■■■
For partial ring
- 3 *i
143*0
■ ' ■
iso struct.
-o *5
SnVa
142*8
V m
142*6
142*5
TABLE XLIV.— Aromatic Haloqrn Compounds.
Substance.
Vm.
2»Va.
A.
C.H.F .
101*6
n6*o
— * 4*4
C.HjCl .
114*6
129*1
“ 14*5
C.H.Br .
120*0
135*3
- 15*3
C.H.I .
1307
144*3
-13*6
Mean value .
•
- 14*5
There is associated with these compounds the normal con¬
traction of - 14*5 f° r the ring, from which it follows that the
volumes of chlorine, bromine, and iodine are not very different
from their volumes in paraffin derivatives.
The volume of—
C 6 H 6 - is 96-0 - 3*2 = 92*8
and of C 6 H 4 = is 96*0 - 2 x 3*2 = 89*6.
The halogens in these aromatic compounds can be shown to
have nearly the same volumes as in the aliphatic compounds R - X
Cl sb 114*6 - 92*8 = 21*8
Br = 120*0 - 92*8 = 27*2
I = 1307 - 92*8 » 37*9
In open-chain compounds
Cl = 21-5
Br = 27*0
I = 37*o
* The value of A found was 1*5 for association of Br and iso group. The value
of - o*5 for the iso group should be added, so that the full value for the former con¬
stitutive influence is + 2*0.
+ See Table XLII. The volumes of the bromine atoms ate taken as
[Br] = 270+ A.
Digitized by v^ooQle
94
LIQUID CHEMICAL COMPOUNDS.
There is, however, indication of a slight increase from Cl to I,
possibly due to the unsaturation of benzene.
TABLE XLV.— Other Aromatic Halogen Compounds.
Substance.
2»V a - R.
A.
1.: 4 C 6 H 4 (CH,)C1 .
I 35’3
137*3
-2*0
C fl H 5 . CH a Cl .
I33’8
136-5
-27
C fl H 5 . CHC1 2 .
1547
154-6
—
1: 2 C a H 4 (.CH 8 )Br .
I42-I
142-7
—
Some of these results are unusual, since o compounds usually
possess contractions and p compounds do not, but the data are not
sufficient to enable us to account for them. Compounds of the
type of o y nty and p dihalogen benzene derivatives have not yet
received attention.
The Gradual Chlorination of the Benzene Molecule.
This very interesting question can be studied owing to the
work of Jungfleisch. 88
The volume of C^H^ is 96*0, and every atom of hydrogen
subtracted involves a loss in volume of 3*2.
Cl
\/
ci
D
Cl
Cl
/\
\/
Cl
vCl
Monochlorbenzene , C 6 H 5 C1. Vm 114*6 b.p. 132
(Unsymmetrical or odd)
Cl «. C 6 H b C 1 - C 6 H 6 = 114-6 - 92-8 = 21-8.
1: 4 Dichlorbenzene , C 6 H 4 Cl a . Vm 130*9 b.p. 172
(Symmetrical or even)
Cl 2 = 130*9 - 89 6 = 41*3.
1:2:4 Trichlorbenzene , C 6 H 3 C1 8 . Vm 149*1 b.p. 213
(Unsymmetrical or odd)
Cl s = 149*1 - 86*4 =» 627.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING THE HALOGENS . 95
1:214:5 Tetrachlorbenzene , C 6 H 2 C 1 4 . Vm 164*8 b.p. 244
(Symmetrical or even)
Cl 4 = 164*8 - 83*2 = 8i*6.
1:2:3:415 Pentachlorbenzene , C 6 HC 1 5 . Vm 183*9 b.p. 276
(Unsymmetrical or odd)
Cl 6 = 183*9 - 80 0 = 103*9.
Hexachlorbenzeru , C 6 C1 6 . Vm 200*0 b.p. 326
(Symmetrical or even)
Cl 6 = 200*0 76*8 a 123*2.
The average volume of chlorine is 20*0 in all the compounds.
A close study of the results, however, shows that the chlorine
atoms are not equal in volume, but that each pair involves a loss
of about 2*3 or 2*5. 34
TABLE XLVI.
Substance.
Monochlorbenzene
c,h 5 ci
V*ci.
21*8
A.
(Unsymm. or odd).
/-Dichlorbenzene
C.H.CIj
4 i *3
I 9‘5
(Symm. or even).
Trichlorbenzene
C.H.Cl,
62*7
21*4
(Unsymm. or odd).
Tetrachlorbenzene
C.HjCl.
8i*6
18*9
(Symm. or even).
Pentachlorbenzene
C e HCl„
103*9
22*3
(Unsymm. or odd).
Hexachlorbenzene
C.C 1 ,
123*2
19*3
(Symm. or even).
There is thus a distinction between the Cl atom on one side
of the molecule, and the Cl atom in the para position with
reference to it, or the two atoms on opposite sides of the ring
involve a contraction of 2*6 on the average.
Thus we have
1st Chlorine Cl*
2nd Chlorine Cl 11
3rd Chlorine Cl* 1 *
4th Chlorine Cl iv
5th Chlorine Cl v .
6th Chlorine Cl vi
Mean values
rABLE XLVII.
Normal.
Abnormal.
21*8
—
—
19*5
21*4
—
—
18*9
22*3
—
—
x 9*3
21*8
19*2
A = 2*6
Digitized by v^ooQle
96
LIQUID CHEMICAL COMPOUNDS ,.
The smaller values apply to the chlorine atoms marked
in the following diagram.
\
\
and the apparent result is that one-half of the chlorine atoms
differ from the other half.
Digitized by v^ooQle
CHAPTER IV.
THE MOLECULAR VOLUMES OF ORGANIC COMPOUNDS
CONTAINING OXYGEN.
The Volumes of Oxygen.
KOPP (loc. cit.) recognized the existence of two different values
for oxygen—
Hydroxyl O' 7*8
Carboxyl O" 12*2.
The numbers are not very different from those now advocated,
but it is necessaiy frequently to suppose that the volume of
oxygen is not always one of those just indicated. Owing per¬
haps to constitutive influences several other values are also found
necessary.
The volume of free oxygen at the boiling-point is, according to
Olszewski, Ber. 17 ref. 198, about 14*0 density (it 10 to IT 37).
V (O: O) = 28-0.
From the critical data,
Tq = 155. Pc = 50 atmos. b.p. = 92,
we find by D. Berthelot’s formula—
v _ M iriTo*
D “ P c (2Po - T)
24*4 for O a .
The atomic volume of free oxygen will be found to be
greater than that of any of its combined values.
Compounds of Oxygen.
Among the simple compounds of oxygen is water H a O,
which, contraiy to expectation, finds its place at the head of the
symmetrical ether series.
H-O-H CH3-O-CH3
Vm(HjO) = 18 9 [O] = 18*9 - 2 x 37 = 11*5
Hydrogen dioxide H - O = O - H Vm = 23*4
2[0 iv ] = 23*4 - 7*4 = i6’o.
Thus [O iv ] = 8-o t
97 7
Digitized by v^ooQle
9 8
LIQUID CHEMICAL COMPOUNDS.
This compound seems to contain two hydroxyl groups—a
fact which is deduced from the diminished volume of oxygen.
A similar diminution is also seen in the alcohol series.
TABLE XLVIII.— The Alcohols C»H a »-fiO'. See Le Bas, Ref. 85
Compound.
V m .
A.
V*.
The paraffins.
C 3 H 7 OH
8i*4
7*4
74*0
c,H 8 .
c 4 h 9 oh
102*1
6*i
96*0
c 4 h 10 .
c 6 h u oh . .
1237
5*9
117*8
C,H U .
C 6 H 18 OH . .
146*4
6*5
139*9
c,h m .
c 7 h 15 oh . .
168*7
6*1
162*6
c,h m .
Mean .
6*4
The volumes of the simple compounds methyl and ethyl
alcohols are somewhat different from the volumes they would
possess, if they were strictly comparable with the succeeding
members of the series. This feature seems to be common to all
homologous series both in the case of molecular volumes and for
numerous other physical properties also.
Alcohol.
v w .
2«V*.
A.
CH S . OH
42*8
36*0
+ 6*8
C 2 H 8 . OH
62*1
58*2
+ 3*9
C 8 H 7 . OH
81*4
81*4
—
The effect of adding the group CH 2 a number of times, is
similar to that observed when progressively large atoms like H,
Cl, Br, and I are added to the radicle CH 8 , as already shown.
This effect, in a wide sense, may be considered to be due to
differences in complexity, or to increase in volume, and it prob¬
ably indicates the existence of differences between the internal
forces of affinity in the different compounds. At any rate it
shows that there are such differences in the distribution of force
in the various compounds, that they are not strictly comparable
at the boiling-point. These features repiesent slight differences
in constitution existing.
We have supposed that the relation
[C] = 4 [H]
holds in the paraffin series.
Digitized by
Google
ORGANIC COMPOUNDS CONTAINING OXYGEN,
So there is every reason to suppose that the relation
[O'] = 2[H]
is true in the alcohol series. The evidence is as follows :—
Propyl alcohol C S H 7 0 H 81*46 = 11 x 7*405
Allyl „ C 3 H 6 OH 74*08 = 10 x 7*408
A for H a = 7*38
If we consider 2[H] to represent a volume of 7*4, then
hydroxyl oxygen also possesses this volume.
Whether this is exactly so or not, the utilization of this
relation enables us to find the variation in the atomic volumes in
series of compounds like that of the alcohols. In calculations it
is better to use the value [O'] = 6*4.
The Branched Hydrocarbon Chain.
TABLE XLIX.
Normal Compounds.
Vm.
A.
v,».
Branched Chain
Compounds.
ch 3
CH 3 . CH a . CH 2 . OH .
81-46 (Sch.)
-0*54
80*92 (Sch.)
ch 8
-CH(OH).
+ 0*49
82-97 (Z)
CH 3 . CH a . CH a . CH a . OH
101*79 (Sch.)
101*86 (Sch.)
CH„
102*11 (Z)
ch 3
CH . CH a (OH).
ch 3 i
0
p
w
+
102*8 (Th.)
CH,
C(OH).
CH,
CH 3 . CH a . CH a . CH a . OH
123 6 (Z)
-0*50
123*1 (Sch.)
ch 3
123*6 (Th.)
ch 3
CH . (CH 2 ) 2 OH.
ch 3
!
- 2*20
121*4 (Th.)
ch 3
,C.(OH)CH 2 .CH 3 .
ch 3 .
• CH a ]
- 1*60
122*0 (Th.)
ch 3
VCH . CH a OH.
CH 3 . (CH a ) 6 CH a (OH) .
190-9 (Z)
+ 0*6
191*5 (Sch.)
ch 3
. CH(OH)C 8 H 13 .
These results are most contradictory, a fact which is rendered
more evident by the observation that the data due to different
observers sometimes show positive, and sometimes negative
differences.
It follows, that no general rules can be drawn up which would
indicate the connexion between these differences and the position
7 *
Digitized by v^ooQle
IOO LIQUID CHEMICAL COMPOUNDS .
of the OH group in the chain, or its position relative to the iso
group.
It is at present difficult to understand the reason for such
differences.
The butyl alcohols of Thorpe indicate a close relationship
between the position of the (OH) group and that of the iso group.
In these instances we are able to state that the minus differ¬
ence is greater the nearer the (OH) group is to the iso group.
The following consideration emphasizes the difficulties at¬
tending a satisfactory explanation of the above results. This is,
that there are possibly three overlapping constitutive effects
which, moreover, may not be quite independent:—
(a) The effect of the branched chain.
( b ) The position and influence of the (OH) group.
(c) The influence of the (OH) group on the iso group.
A re-determination of all the values by one observer, and a
more extended body of data would doubtless clear up these
difficulties.
Partial or Incomplete Rings . 86
a/? and ay Compounds.
TABLE L.
Compound.
Vm.
2„V a .
A.
Ethylene
Glycol CH a (OH) . CH a (OH)
64-9*
66*4
-i *5
Propylene
„ CH 8 . CH(OH) . CH a (OH)
85-4
88*4
-3*o
Trimethylene „ CH,(OH). CH, . CH a (OH)
8,f2
ft
“ 4*2
These compounds are analogous to the a /3 and ay halogen
compounds, and the values of A are :—
CH3-CH
CH 2
/
ch 2
\
ch 2 -
CK
ch 2 —
A — — 3»i
^ = —4.3
CH 3 — CH
— Cl"
CH 2 —
A = — 3.0
Brx
ch 2 —
\
/
1
1
/
/
ch 2
\
Br''
ch 2 -
-OH n
OH'
-OH >
•OH''
^ = -4.2
* This value due to Ramsay is probably too large, thus making the value of A
too small.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN ioi
The contractions are thus due to particular configurations of
the molecules which have been already noticed. They have been
named partial or incomplete rings.
The following conclusions are drawn:—
(a) The a/8 are smaller than the ad compounds by about
3 units.
( b ) The ay are still smaller than the ad compounds by about
4 units.
The following compounds show how the atom : O contrasts
with the halogens, and with the (OH) group in respect of its
effect in the direction indicated.
(i) Acetyl chloride.
(2) Triacetyl chloride.
Cl Cl
Cl 0
Cl
CH S — i or CH 3 — (! = O
1 II
ci—C—C or
0
11
°\
— 0
1
0
II
O
i, t,
V» 74*0 (Thorpe). 87
CH* 26*0
C 14*8
\m 125*5 (Thorpe).
2C
29*6
4 C1
88*4
Cl 22*1
: 0 n*o
: O
11*0
—
5nV a
129*0
2nVa 73*9
Vm
125-5
A
-3*5
We see that the atom : O in
opposition with
a halogen atom
does not produce the constitutive effect indicated, but that two
halogens in opposition are capable of doing so.
It is evident that in triacetyl chloride we have two adjacent
chlorine atoms which produce the normal contraction. The
association of the atom : O with Cl does not produce any very
appreciable contraction, either owing to the small attraction
between the two atoms, or because the double linking of the
oxygen causes it to lack mobility.
The formula for the compound in consequence might be—
Oi
I
1
pi _p
_ P = A
Vtl \J
— u = u
1
il
1
Cl
Digitized by
Google
102 LIQUID CHEMICAL COMPOUNDS.
(3) Weger 88 has studied the volume of an a/8 compound of
the same nature.
afi dibrompropyl alcohol CH 2 Br . CHBr . CH 2 OH.
b.p. 219 0 V m 124*3.
Volume of CH 8 . CH 2 . CH 2 (OH)
81-5
less H 2
- 7‘4
plus 2 Br
74*1
56*6
2 nV a
130*7
Vm
124*3
A
-6-4
The formulae which might conceivably apply are:—
Br-CHo
i
(a) Br ^
i
t
ch 2 — o\i' J
(b)
ch 2 -
- Br \
1
)
/
CH-
-Br/
HO — CH. 2
A = - 3-0
— 3-0
CH 2
/
(c) Br-CH
CHa
A = - 4 • 5
CH 2 - 0 H\
\
i
/
CII-Br(^
\
\
I
/
CH 2 -Br - /
— 6.0 probably
The first two, and possibly the third, are excluded by reason
of the magnitude of the contraction. Compounds of the type
expressed by the fourth formula have not been directly studied.
These include the trichlor, brom hydrins. In order to ascertain
the contractions to which these are subject, their volumes maybe
calculated by the formula already given.
Weger gives for a /3 dibrompropyl alcohol. CH 2 Br . CH 2 Bt . CH 2 OH.
do 2*1682 b.p. 219 0
dsi9 17535
Thus C = 0*532.
For trichlorhydrin, CH 2 C 1 . CHC 1 . CH 2 C 1
we find d 16 1*3980 (Perkin) b.p. 156*0.
: - Assuming the value C = 0*532, we find
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. 103
CH 2
CH-
1
ch 2
CK
Cl
Cl
1
/
N \
J
Vm 124*2
C 8 H 5 17 x 37*0 62*9
3CI 66*3
2nVa = 129*2
Vm = 124*2
A = -5*0
For tribromhydrin, CH 8 Br - CHBr - CH 2 Br.
CH 2
CH-
CH 2
Br
\
•Br
/
•Br
b.p. 220°
Vm
di 5
sH
3 Br
ZnV a =
Vg =
A =
2*4134 (Perkin)
141*1
62*9
84*0
146*9
141*1
- 5*8
These trihydrins are evidently similarly constituted to the
parent compound glycerol, and we are thus led to suppose that
Weger’s compound possesses a similar constitution, by reason ot
its contraction.
CH 2 -OH>^
)
I
CH- 0 H(^
CH 2 -OH''
Glycerol.
A few more of these compounds may now be studied,
a dichlorhydrin CH 2 C 1 . CH(OH) . CH 2 C 1
CjHjCl.O
d„ 1-367 b.p. 174° C =
0*535
Vm =■ II2-7
ch 2 -
- CK 3CH ?
66*6
\ 2C1
44*4
1
1 -O-
/
7*4
CH -
ii8*4
I
\ By form.
1127
1
ch 2 -
-C\J A
= -57
The 0 dichlorhydrin CH 2 (OH) . CHC 1 . CH 2 C 1 has a slightly larger volume
113*5, from which we obtain the difference A = - 5*1.
CH 2 -
-Br\
|
\
1
Br 7
CH 2 -
1
1
-OH^
Dibrompropyl alcohol.
Digitized by v^ooQle
104
LIQUID CHEMICAL COMPOUNDS,
i„>
iff,!
,OH
Glycide - epihydrin alcohol C,H 6 O a .
The value of C from epichlorohydrin is 0*46.
do 1*165 b.p. 162 0 V m = 74*4
3CHj 66*6
14-8 (2 X 7-4)
20
5nVa
fey form.
81*4
74*4
A fofc ring = - 7*0
A compound of this type has been directly investigated.
Epichlorohydrin.
CH,\
1 \n
V m = 87*3 (Thorpe)
CH /
C 8 H 6 63*0
Cl 22*2
CHjCl
O 7*4
2 nVa = 92*6
By Obs. = 87*3
A for ring = -5*3
These two compounds thus possess three-membered rings.
It thus happens that the value of A for these ring compounds is
similar to those which are constituted the same as glycerol.
This is probably a coincidence only. It is, however, possible
that part of this value may be due to the attachment of the
group CH 2 C 1 to the ring, since a contraction occurs in the com¬
pound C c H 6 CH 2 C 1 for this reason.
In conclusion we see from a study of their molecular volumes
that the compounds have without doubt the formulae which have
been given to them.
Aromatic Compounds.
TABLE LI.— The Phenols.
Compounds.
Vm.
A.
Vm.
The Hydrocarbons.
C 6 H 5 OH . . .
101*9
5*9
96*0
C 6 H 6 Benzene.
*C,H<(CH,)(OH). .
123*8
5*6
118*2
C 6 H b (CH s ) Toluene.
C.HjCHjOH .
1237
5*5
»»
Mean
= 5*6
Digitized by v^ooQle
ORGANIC COMP 0 t/NDS CONTAINING OXYGEN 105
The volume of - 0 - in the phenols is on the average 5*6,
and thus is 6*4 - 5*6 = o*8 of a unit smaller than in ordinary
alcohols.
This difference indicates the existence of a special constitutive
effect caused by the substitution of the aromatic nucleus for the
alkyl group R, when the group (OH) is present. For C 10 H u O
two values are given.
(а) Ordinary thymol Vm 189*3 (Pinette).
(б) Carvol (unspecified) 190*7 (Schiff).
From the formula, which is that of thymol,
we get the following calculated volume:—
Cymene 184*6
less H - 3*2
181*4
plus OH 9*3 (5*6 + 37)
2«Va 190*7
This is exactly the value given by Schiff It will be seen,
however, that the (OH) group occupies the ortho position with
reference to the isopropyl group, and this should occasion a con¬
traction of about 2 units. The normal volume thus is 1887
approximately, which is similar to the volume given by
Pinette.
Moreover any other compound of formula C 10 H 14 O would
still be subject to a similar contraction, for an OH group in the
nucleus would necessarily occupy the ortho position with refer¬
ence to either the C 8 H 7 or the OH group. Thus the volume
of any of these compounds would be in the neighbourhood of
189*0.
There is no reason to suppose that Schiffs value is due to
experimental error, so that we are compelled to find some other
reason for the larger volume.
Remembering that the latter corresponds to a compound in
which the paraffin and hydroxyl groups are independent, that is
to 3»Va, only allowing for ring structure, we see that the ex¬
planation may lie in the fact that the groups are opposed. The
structure of Pinette’s compound would be “ adjacent ”.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 107
the value of Sn V a > and so results in one very nearly equal to
that shown by V m .
It is to be supposed that compounds like menthol C 10 H 20 O
should agree with the characteristics of the class to which it belongs.
Thujamenthol, 0*9015 b.p. 212 0*7389 C = 0*554
V m = 211*1 (thymol)
From thymol we find—
Cymene H 14 184*5
O' 5*6
190*1
less for ortho struct. - 2*0
188*1
6H 23*8
2nV a = 211*9
Vm = 211*1 by form.
There is thus agreement with the formula—
CH 2 —CH-OH
/ \ / CH
CH,—CH CH—CH<
\ / XCH
CH,—CH,
3
3
This compound possesses a contraction of - 15*0 for ring
structure, and a contraction of - 2*o for ortho structure, as for
thymol.
CH
CH
Dihydrocarveol c,.h 18 o
djQ 0*927 b.p. 225 0
2 CH 2 —CH—OH
\ / \
C—CH CH—CH.
/ \ /
3 ch 2 —ch 2
C = 0*550
^386 °*74 I
Vm 207*8
JSnVa 222*0
A - 14*2 for ring.
This at once enables us to place dihydrocarveol and similar
compounds in their particular class—the menthan terpene alcohols.
A very interesting compound is—
Cineol C 10 H 18 O.
CH a
CH 2
CH.
\
CH
\
CH,
1
CH*
CH*
/
H
This substance has the following formula
ascribed to it:—
C = 0*46 as is usual for ring compounds,
b.p. 176 d 16 0*923 d 176 0*792
Vm 194* 0
2»V a 222*0
A = - 28*0 (for ring).
Digitized by v^ooQle
io8
LIQUID CHEMICAL COMPOUNDS .
Cineol thus belongs to the bridged ring class.
A compound of a similar character which has been specially
studied is
Borneol C 10 H 18 O.
CH a -CH-
CH,—C—CH a
CH,- c!-
<^H,
■CH,
-CH(OH)
V w 190-5
%Na 222-0
A = - 31*5 for ring.
The value of A for these compounds is about twice that for a
single six-membered ring.
It seems probable that minor overlapping effects may disturb
the result somewhat.
Thus 2 x - 15-0 *= - 30*0 for ring structure,
and - x«5 for ortho position of CH 8 and OH.
In accord with this, we find that camphor, a-CO compound,
possesses a contraction of - 30 0 only.
The Ethers.
R—O—R'.
Corresponding to the alkyl halogen compounds (halogen
monovalent)
R—Cl, R—Br, R—I
we find the ethers ( divalent)
R—O—R and R—O—R^
which involve the combination
C—o—C.
These give results which are different to those characteristic
of the combination which has just been studied, viz.:—
C_o—H.
The following isomers of empirical formula C 6 H 14 0 exist—
c 3 h 7 —o—c 8 h 7 C 4 H 9 —O—C 9 H 6 C 6 H u —O—CH 8
¥ V
Symmetrical. Unsymmetrical.
Vm 151*3 I 50*3 149*5
C 6 H l8 —O—H
V m 1462.
In the above compounds, there is a progressive change in
volume from the symmetrical to the unsymmetrical types, and it
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 109
becomes smaller as the oxygen atom approaches the end of the
chain.
An analogous variation is found in the case of the boiling-
points, but in the opposite direction.
(1) (2) (3) ( 4 )
b.p. 907 91*4 — 157-0
The last step of the change in boiling-point, that is, from the
methyl ethers to the isomeric alcohols, is usually very great—about
70° C.
This corresponds to a volume change of 5 units. The follow¬
ing comparisons give the volumes of ethereal oxygen. The
values of this atom are found by subtracting the volumes of the
paraffins from those of the corresponding ethers.
TABLE LII.
The Methyl Ethers .
The Symmetrical Ethers .
The Paraffins . |
Compound.
v w .
>0.
Compound.
v m .
>0.
C»Hjw + 2.
v«.
H— 0 —H
18*9
n *5
cX
[7*4]
CH S —O—CH,
627
107
[52-0]
C a H 5 —0—CH 3
84*2
10*2
C,H„
[74-0]
CjH 6 —O—CjHj
106*4
10*4
C,H l0
96*0
CjH 7 — 0—C H 8
105*9
9*9
c 4 h 9 —0—ch 8
C B H n —0—CH S
127*5
9*7
C.H.,
117*8
149*5
9*6
CjH,—O—C,Hj
I 5 1 *3
n*4
c„h 14
139*9
c 7 h 16
l62*6
C*H,—O—C 4 H„
197-8
n*5
c 8 h i8
l86 *2
The mean value of ^>0 in the methyl ether series, is 9 9,
and in the symmetrical series n*o. Also, there are similar
variations in the two series, viz. first a decrease, and subsequently
a gradual increase. These changes are, however, comparatively
small.
The Position and Nature of Liquid Water.
It is very remarkable that, in accord with its simple formula,
water should stand at the head of the symmetrical ether series—
and is, in fact, its vanishing point.
In this it is analogous to hydrogen (H a ) which is the vanish¬
ing point of the hydrocarbon series. In contrast to unsymmetri-
Digitized by v^ooQle
no
LIQUID CHEMICAL COMPOUNDS.
cal methyl alcohol, water is a symmetrical compound, and there
is a corresponding difference in the volumes of O in the two
compounds.
H—o—H
CH S —O—CH 3
The symmetrical ether
series.
- H,
CH 3 —O—H CH 4
The alcohol The paraffin
series. series.
The formula apparently indicates the existence of a hydroxyl
group, but it is more in accord with the formulae for the sym¬
metrical ethers.
As regards its volume, which is 187, we may suppose that it
is made up as follows:—
[H 2 0 ] = 2[H] + [O] = 2 x 37 + n*3 = 187.
The volume of oxygen ^>0 is thus 11 3.
This is considerably larger than that of hydroxyl oxygen as
found in the alcohols.
[C 6 H 13 OH] = [QHjJ + [O] = 139-9 + 6-5 = 146-4.
The volume of O' is thus 6*5, a number which is quite
different from its former value. The volume of oxygen in water,
however, agrees with the idea that on the whole it is ethereal in
character. An examination of dimethyl ether shows this.
[(CH 3 ) a O] - 2[CH 3 -] + [> 0 ] = 52-0 + 107 = 62-7.
This value 107 is not very different from that obtained from an
examination of water.
These calculations show that water is not a hydroxyl com¬
pound, but one of the symmetrical ethers.
It has been known for some time that the simple formula
H 2 0 , whilst doubtless representing one particular type of mole¬
cule, does not express the nature of water as a whole. Liquid
water is an associated substance, that is, it is made up of mole¬
cules which are polymers of H 2 0 . Its formula might be ex¬
pressed by (H 2 0 )at, where x is a number which is greater than
unity. It is, moreover, known that the value of x cannot be
expressed by a simple number, but water is a complex of many
types of molecule.
Armstrong 89 has endeavoured to show that these fall into two
classes.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. hi
(a) Those which are similar in configuration to the paraffin or
open-chain series —in fact to the normal alcohols.
H
H
/
h 2 o h
/
\/
h 2 o
0 H
etc.
\
/\/
OH
H 0
or 0 H 3 —OH
OHg— 0 H 2 —OH
which are similar to
CHg—OH
CH 3 —CH 2 —OH
and so on.
(U) The hydrones which
are similar to ethylene and the poly ■
methylenes .
H,
H a H 2
H 2 0 « 0 H 2
6
6—6
/\
1 1
H a . O O . H,
0—0
Dihydrone.
Trihydrone.
H„ H a
Tetrahydrone.
h 2
h 2 h 2
CH 3 = ch 2
c
C—C
/\
1 1
H a . c — c . h 2
c—c
Dimethylene.
Trimethylene.
H a H,
T etramethylene.
These formulae are consequent upon the assumption of
quadrivalent functions by oxygen.
It may be shown, by means of molecular volumes, how far
this hypothesis is justified at the boiling-point.
We do not know the volume of quadrivalent oxygen ex¬
cept as indicated by the single compound hydrogen-dioxide
H- 0 = 0 - H.
The volume of - O = is found to be 8*o.
Since, with the exception of water, all of the above formulae
involve quadrivalent oxygen, the average volume cannot be
greater than 15*4, if we consider compounds of class (a) alone
and of complexity higher than (H 2 0 ) 2 .
On the other hand, if we consider the influence of the
hydrones, we come to the conclusion that the average volume
would be considerably less if these existed in any large proportion,
because ring compounds are subject to considerable contractions.
Digitized by v^ooQle
I I 2
LIQUID CHEMICAL COMPOUNDS .
We are thus left with the simple molecule H s O and the more
complex ones, OH 2 = OH 2 and OH 8 - OH.
The volume of OH 2 = OH 2 would be 35-5, comparing it with
ethylene. This gives an average value of 17*7.
The volume of OH 8 . OH would be 36*2, or an average of
18-1 for H 2 0 .
If, however, we remember that the tendency of H is to increase
the volumes of the simple compounds in which it occurs, as for
example in CH 4 , QH^, HC 1 and the initial members of most
series, the volume of simple H a O at the boiling-point might be
20 0 or even more.
On the whole, we think that at the boiling-point liquid water
might consist mainly of
H a O, OH 8 . OH and OH 2 = OH 2
with perhaps smaller proportions of compounds of complexity
(H 2 0 ) 3 . It is not, however, probable that the other more com¬
plex compounds could occur, especially those which involve ring
structure.
In accord with this we find that most physical properties
show an association factor of between 2 and 3.
Physical Property.
Association Factor.
Capillarity (Ramsay and Shields)
26
Mol. Cohesion (Walden)
1*98
Volume (Traube) ....
30
Fluidity (Bingham) ....
2*2
Mean
2*4
Certain other ethereal compounds are interesting.
Ethidene dimethyl ether C 2 H 4 (OCH 8 ) 2
CH 8 . CH
/OCH 8
\ 0 CH,
V m =
(CH,) a CH . CH,
V*
II 2'3
96*5
II 2’3
2[0] = 157
[> 0 ] - 7-85
This volume of the above ethereal oxygen is similar to what
we find in QH,—O—CH 3 , but considerably less than in
C S H S —O—CH a .
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 113
ether C 2 H 4 (OC 2 H 5 ) 2
Vm = 160*2
(C 2 H 6 ) 3 CH . CH S 139-0
C 2 H 4 . (OC 2 H b ) 2 160*2
2 >0 21*2
>0 10*6
The volume of the ethereal oxygen is similar to what we find
in C tf H 5 — 0 —C 2 H 6 , and is more in accord with the number
found in ethers of the type R—O—R'.
The above compounds thus show agreement with the ethers
of the monohydric alcohols, in that the methyl group causes a
depression in volume as compared with the volume when the
ethyl group is present.
The explanation of the particular value found is not quite
easy. The molecules are of the branched chain type of structure,
the effect of which in this class of compound has not been studied
sufficiently. In compounds with a single —OCH 8 group, the
depression is (io*6 - 9 6) ro, as compared with compounds with
those having a single —OC 2 H 5 group.
In those with two —OCH 3 groups, the total depression is
2 (io-6 - 7-85) = 2 x 275. On the whole, there is an outstanding
difference of 3*50.
The Effect of the Addition of the Homologous Increment
CH 2 on the Volumes of the Ethers.
The aliphatic ethers (see Dobriner, ref. 17) may be divided
into series, e.g. that of the methyl ethers CH 8 —O—CH 8 , the
ethyl ethers C 2 H 5 —O—C 2 H 5 , etc., each of which symmetrical
compounds stands at the head of a series. The methyl CH 8 ,
ethyl C 2 H 5 , etc., groups will appear on one side throughout the
series, a radicle of variable complexity appearing on the other side.
It has been shown that the volumes of the ethers and those
of certain esters are nearly the same, and that both series may be
represented by a similar number of hydrogen equivalents W.
The ethers C»H 2n + 2O show values of W which are equal to
6 n + 5, because [> 0 ] = [3H] and [C] = 4 [H].
The esters C»H 2w O also show values of W equal to 6 n + 5,
because O' = 2H and O' = 3H.
If we know the molecular volume of a compound and the
8
Ethidene diethyl
OC 2 H 5
CH„. CH/
\oc 3 h 5
Digitized by v^ooQle
Values of v /w
LIQUID CHEMICAL COMPOUNDS.
114
number of hydrogen equivalents, the ratio V/W gives the volume
of a single hydrogen equivalent This enables us to find the
variation in the volumes of H, and thus of every atom, throughout
the homologous series.
TABLE LIII. —The Aliphatic Ethers.
R—1
0—R'.
The Dimethyl series.
The Diethyl series.
w.
Compound.
V m .
V/W.
W.
Compound.
Vm.
v/w.
17
CHj—O—CHo
62*6
3*680
29
c 2 h 5 —0—c 8 h 5
106*2
3-664
23
CHj—O—C 2 H b
84*2
3*660
35
c 9 h 5 —0—c 8 h 7
128*1
3-659
29
CHj—O—C 8 H 7
105*1
3*624
4i
c 2 h 6 —0—c 4 h 9
150*3
3-666
35
CHj—O—C 4 H 9
127*5
3*643
—
—
—
_
_
__
59
c 9 h b —0—c 7 h 16
221*3
3751
53
CHj—O—C 7 H 15
195*0
3*680
63
c 9 h 6 — 0—C 8 H 17
247*2
3-803
59
CHj—O—C 8 H 17
220*2
3*732
Dipropyl series.
Dibutyl series.
W.
Compound.
M.V.
V/W.
W.
Compound.
M.V.
V/W.
4i
CjHt-O-CjH,
151*3
3*689
53
C 4 H 9 —0—C 4 H 9
197*7
3-728
47
CjHt-O-C.H,
173*2
3*686
—
—
—
_
_
_
7i
c 4 h 9 —0—c 7 h 15
272*0
3-831
65
C s H 7 —O—C 7 H 15 246*1
3786
77
C 4 H 9 —0—C 8 H 17
297*4
3-862
7i
^8^7-O-C 8 Hj 7 273*0
3-845
20 30 40 50 60 70 80
Number oP H Equivalents.
Fig. 3*
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 115
The above calculations show that, except at the beginning
of series, there is a general increase in the value of V/W.
Normally it should eventually be abouj 0*30 for each addition
of CH 2 . There are signs that this is the case, if we depend
on the evidence of those series which do not include C 7 H 15 or
c 8 h 17 .
The differences in the latter cases are over 0 50, which is
double to that ordinarily found. This makes the slope of the
curves for these complex compounds steeper than usual, e.g.
as shown by the line C 38 C 37 , etc., in the diagram.
There is thus an apparent absence of continuity in the values
of V/W between the simple and the more complex compounds.
An examination of the C 5 H n and C c H 13 ethers would show
whether this is real or not, and in the latter case, how the two are
connected. An attempt has been made to show this connection
in the curve for the ethyl ethers C 22 . . . C 28 . Excluding the
methyl ether curve, which is different from the others, the curve
just mentioned probably resembles in form those for other series
as e.g. the normal paraffin series.
Thus, we find a nearly horizontal portion like ^ 22 ^ 23^24 which
includes a minimum. This minimum is very pronounced in the
methyl ether curve, is still apparent in the next succeeding,
and then practically disappears in the others.
Subsequently, we find an ascending portion which appar¬
ently is partly coincident with the ascending portion of the
methyl ether curve C 14 . . . C 18 . This finally leaves the straight
line, and the slope becomes steeper, as shown by the parts
^ 27 ^ 28 * C 37 C 38 .
A remarkable fact, however, is, that the compound C 17 , instead
of finding its position on the line at a point marked O, is situated
considerably below this, viz. V/W 3708 - 3*680 = 0*28. If,
however, we join the points C 17 and C 18 , we get a line which
is parallel with the steeper portions of the heptyl and octyl
curves, although C 17 is out of position. This is all the more
remarkable.
It follows from the fact that the gradients of C 27 C 28 , and
C 87 C 38 are much steeper than has been found usual, and also that
C17C18 has a similar gradient, although C l7 is out of position, that
the molecular volumes of these and perhaps other compounds, are
liable to vary from the expected values, from causes at present
8 *
Digitized by v^ooQle
n6
LIQUID CHEMICAL COMPOUNDS,
unknown. These variations, however, do not seem to be due to
experimental errors, nor to the existence of small quantities
of accidental impurities. The whole question of the causes of
the divergences of molecular volumes from those expected,
and the causes of variation in the values given by various
observers, is a difficult question, but well worth study. It is
probable that unlooked for features might be discovered, either
as applying to the molecules themselves, or to the characteristics
of liquids.
The Symmetrical and Unsymmetrical Ethers.
Ignoring the divergent values, and depending on the principle
of continuity, we obtain the following curves shown in the
diagram (Fig. 4).
I, The curve for the symmetrical compounds,
II, The curve for the unsymmetrical compounds,
III, The curve for the methyl series,
I. and II. are similar in form , which would lead us to conclude
that the form is dependent on complexity , or the number of C
atoms in the chain, and not on their character, symmetrical or
otherwise.
Their position is, however, dependent on the relative number of
C atoms on each side of the typical oxygen atom i.e. on
the degree of symmetry.
Curve I. represents compounds of type An = o, or the
perfectly symmetrical.
Curve II. represents compounds of type An = 1, or com¬
pounds less symmetrical by one C atom.
This difference of one C in the radicles on either side of O is
shown by a displacement of the curve in, say, the direction of
increasing complexity, that is, there is a relative decrease in the
value of V/W for want of symmetry.
In the case of the simpler compounds, an additional divergence
from perfect symmetry is shown by an additional diminution in
volume, but higher up this apparently vanishes, the compounds
diverging more than An = 1 from perfect symmetry, are nearly, if
not entirely, of the same volume as those for which An = 1
is true.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN . 117
It is to be observed that the most symmetrical compounds,
that is, those which stand at the head of each series, possess con¬
siderably larger volumes than those which might be expected.
This gives that peculiar feature to the curves which may be
observed from an inspection of the diagram, namely, that portion
which occurs before the minimum, and which is marked by
decreasing values of V/W. Since this characteristic also applies
to the paraffin series, and as a matter of fact to all series, we see
the analogy of these more symmetrical compounds with the first
members of the other series. Dimethyl oxide is, for instance, a
symmetrical compound, but it is also the first member of the
methyl ether series, and corresponds to methane, methyl alcohol,
methyl bromide. It seems to follow that the particular feature
of symmetry, which is a characteristic of dimethyl ether, is that
which also distinguishes the initial members of the other series,
and which influences their volumes. Whether the cause be
simple or complex, as indeed is probable, the operating influence
is no doubt the general shape of the molecule and possibly
its complexity or length of chain. For the present, limiting
our view to the ether series, we distinguish two classes of com¬
pounds—
{a) the symmetrical ;
and {b) the less symmetrical ,
that is, we have to distinguish those in which R- R' = 0 from
those in which R-R' = CH 2 . The special feature applying to
the latter is apparently confined to the ether series and is ad¬
ditional to that which may apply to the compounds R- R' = 0 ,
as distinguished from those marked by the characteristic
R - R' = «CH 2 , and which includes all the members of the series
which are less symmetrical than the absolutely symmetrical com¬
pounds.
Certain well marked diminutions are noticed in the com¬
pounds which are most unsymmetrical, i.e. those which contain a
single methyl group, and sometimes, but to a lesser extent,
those containing an ethyl and even propyl group. Whether this
diminution is due to non-symmetry simply, or to the special
influence of the CH 8 , C 2 H 5 , etc., groups, is a very difficult ques¬
tion to answer, and indeed is that one which presses most.
Digitized by
Google
Values oF yto
118
LIQUID CHEMICAL COMPOUNDS.
TABLE LIV.— The Aliphatic Ethers.
R—O—R'.
Symmetrical.
R—R' = 0 .
W. Compound. V w . V/W.
17 CH 3 —O—CH 3 626 3680
29 C 3 H 5 —O—C 2 H 5 106*2 3*664
41 C 3 H 7 —O—C 8 H 7 151*3 3*689
53 C 4 H 9 - 0 -C 4 H 9 197*6 3*728
65 QH11—O—C 5 H u — —
77 C 8 H X3 O C 6 H 13
89 C 7 H 15 —O—C 7 H 16 353 42 3*971
Unsymmetrical.
R—R' = CH a .
W. Compound. V m . V/W.
23 CH 3 —O—C 4 H 5 84*2 3*660
35 C a H 5 —O—C 3 H 7 1281 3*659
47 C 3 H 7 —O—C 4 H 9 173*2 3*686
59 C 4 H 9 —O—C 6 H n — “~
71 C 6 H X1 —O—C g H 13 —
83 C 6 H 13 —O—C 7 H 15 — —
95 C 7 H X5 - 0 -C 8 H 17 377*6 3*975
Number of H Equivalents
Fig. 4.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 119
Formula Representing the Results for Complexity and
Symmetry in the Ethers. See Le Bas, Ref. 40
An attempt has been made to represent the above results by
a formula.
It has been shown that for the n paraffin and other series, the
formula
v m = W{S + (W - 4 o)K }
reproduces the results—
W indicating the number of H equivalents;
S the least or minimum volume of H in the series;
K a number indicating the variation in volume of one H equivalent for an
addition of one H or its equivalent.
The above-mentioned curves require a more complex formula
to indicate them.
I. The effect of complexity is proportional to the cube of the
number of C atoms.
Al = 0r~ 2 ) (l~ I ) x 00101
and
II. The effect of want of symmetry is proportional to the
square of the number of C atoms, the symbol n representing
their number.
As = n 0 x 0-003.
The whole formula is—
M.V. = (6» + 5) {3-664 + - 2 ) (j - *) * o-oio - » 0 x 0-030}
W = 6n + 5 S = 3*664.
TABLE LV. —Calculation of the Volumes of the Symmetrical and
Unsymmetrical Ethers by Means of Formula.
Class.
Compounds.
w.
n .
V/W.
Corr.
for
unsymm.
v/w
calc.
V/W
obs.
V calc.
V obs.
S
CH S .
0 .
CHo
17
2
3664
3-677
62-25
62-5
U
CH„.
0.
C,H»
23
3
3-661
- o-oog
3-652
3-652
84-00
84-0
S
C 2 H b
0
C,H S
29
4
3-664
3-664
3-664
106-25
106*2
U
0
C»H,
35
5
3-67*
- 0-015
3-657
3-659
128*00
128*08
S
c,h 7 .
0
c,h 7
6
3-684
3-684
3-689
151-04
151-27
u
c,h 7 .
0
• c 4 h„
47
7
3705
-0*021
3-684
3-686
173*15
173*25
s
£H,.
0
53
8
3725
3-725
3-73o
197*42
197*72
u
L> 4 H 9
0
C 8 H n
59
9
3752
- 0-027
3-725
3-733
219*77
22035
s
C.H U
.0
•C,H U
65
10
3785
3-785
3-789
246*02
246*28
u
C»H U
.0
c*h m
7i
11
3*823
-0-033
3*790
3-795
269*09
269-44
s
C,Hj,
.0
• C,H 1S
77
12
3-866
3-866
3-867
297*68
297-76
u
c 7 h„
.0
■C,H U
83
13
3*9i4
-0-039
3-875
3-870
321-62
321*21
s
C,H„
.0
■ c 7 h u
89
14
3-967
3-967
3-970
353-06
353*42
u
c 8 h 17
.0
• c 7 h„
95
15
4*025
3-980
3-975
378-1
377-6
The compound (C 8 H 7 ) 2 0 is not given as its volume does not conform to the
rule just given. Young 41 has, however, shown that the b.p. for this compound is
too low. Given a higher number, there would probably be coincidence between the
calculated and theoretical values.
Digitized by v^ooQle
120
LIQUID CHEMICAL COMPOUNDS .
In studying the compounds of the methyl series, either by
means of Table LIII., or better from Fig. 3, we see the effect
of a CH 3 group on the volume of a compound
R—O—CHg
is, generally, to depress the volume. This is so at least down
to the compound in which R = C 3 H 7 , but as R becomes less
complex, the effect is an augmentation of V/W.
V/W
CH 8 . CH a . CH 2 .0 . CH 8
3*624
ch 8 . ch 2 .0. ch 8
3’66o
CH 8 . O . CH 8
3680
This result may be ascribed to a number of causes : (a)
diminishing complexity, (b) increase of symmetry, (c) increase in
the compactness of the molecule. Probably all of these influences
contribute to the result, but they are not independent, and thus
the analysis cannot be made.
An Examination of the Various Constitutive Effects
Associated with Hydrocarbon Chains.
In the following attempt to analyse the conditions which
give to the molecular volumes of the ethers their particular
character, it might seem that we look upon the factors as being
independent one of the other. In reality, it is probable that
they are not independent , so that real analysis is difficult, if not
impossible. Still our view of things may not be without its
advantages and of some interest.
The influences operating on hydrocarbon chains are:—
(a) Complexity with which may be included length of chain.
The effect of this is probably always to augment the atomic
volume. By complexity we do not necessarily mean increase in
the number of atoms, still less, and increase in the molecular
weight, but rather increase in the number of units of volume
(see Fig. 5).
This is, in normal chains, complicated by a second factor.
(b) One which is closely related to compactness of molecular
form .
This is a maximum for the first member of each series—the
methyl compound.
If we take the carbon atom as a centre, it is possible, by
assuming that the shape of methane is a regular tetrahedron, to
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. 121
describe a sphere about this centre so as to include all the H
atoms. These H atoms would be situated on the surface of the
sphere.
In the case of ethane and propane , etc., the corresponding
figures would be gradually elongating ellipsoids of revolution.
Plane projection would show them as circles and ellipses.
This compactness, or rather diminishing compactness, tends
to diminish the atomic volume, but the observable influence of
this factor, extends to only the fourth or fifth member of each
series. At this point the influence of increasing complexity
becomes predominant.
The Influence of Compactness of Molecules.
Unbroken C Chains.
Homogeneous Attachments.
The n Paraffin Series.
f T \ f
(h— c— h]
1
H
v/w 4 - 8.5
No. of atoms 1
in molecular
chain.
H
H— A— OH
k
V/W 4-28
No. of 1
atoms in
chain.
H H H h\
I I I I \
-O^O-O-O-H]
1111 /
H H H H /
Unbroken Chains.
Heterogeneous Attachments.
The Alcohol Series. OH = 3 V/W.
H H
H H H
H H H H
-A— C—OH
I I
1 1 1
H—C—C—C-OH
1 1 1
1 1 1 1
H—C—C—C—C—OH
l l 1 1
H H
H H H
1 1 1 1
H H H H
3*88
3*70
3*647
2
3
4
V/W
No. of 1
atoms in
chain.
The Chloride Series. Cl = 6 V/W.
H
H H
H H H
H H H H
H—t—Cl
H—A—(t—Cl
H—i—A—i—Cl H-
—i—(1 —C—t
A
A A
A U
u u
3-90?
3774
3-668
3-661
Cl
Digitized by
Google
122
LIQUID CHEMICAL COMPOUNDS .
The Bromide Series . Br = 7*5 V/W.
H
H H HHH
H H H H
H—C—Br
H— i—i— Br H—t— i —d—Br
H— A—A—<L- i— Br
k
U iik
u u
V/W 3-84
3780 3*672
3'®49
No. of atoms 1
in chain.
2 3
The Iodide Series. I = 10 V/W.
4
H
1
H H HHH
H H H H
H—i—C— L —C—I
H—C—I
1
II III
H—C—C—I H—C—C—C—I
11 111
1
H
II J
H H HHH
a i. a
V/W 3*770
3730 3*683
3*663
No. of atoms 1
in chain.
2 3
Broken Carbon Chains .
The Ethers.
4
H
H HHH
HHH H
V/W
No. of atoms
in chain.
V/W
No. of atoms
in chain.
H—i—O—A—H H—C—C—0—C—H H—i—i—i—o—i—H
t i H i kkk i
3*680 3*660 3*628
3 4 5
HH HH HHH HH
H _u —O— u —H H—i—it— o —i—C—H
u u ha a
3*664 3*659
5 6
In the above-mentioned compounds the influence at work
is, as it were, a mutual repulsion between the atoms attached to
the same C atom.
Such a mutual repulsion is exerted by H, Cl, Br and I atoms
among themselves, and the effect is a maximum when all the
atoms attached to the carbon are of the same kind (see p. 12 5).
The rule, which may be applied to the different series, is,
that the volume tends to a maximum in a series , as the shape of
the molecule becomes most compact , i.e. when it may be enveloped
by a sphere of revolution . This will naturally be the simplest
member.
When there is reason to believe that mutual attractions are
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 123
possible, the above rule is no longer true. The effect indeed is
exactly the opposite of the above. An example of this is found
among the branched-chain hydrocarbons. The rule is
That the volume tends to a minimum , as the shape of the mole -
cule approximates to that of the sphere .
An example of this attracting influence between groups is
Tetramethyl methane.
Di-isopropyl.
CH,
ch 3 ch 3
11
CH,—C—CH,
1 1
H—C-C—H
1 |
CH,
1 1
ch 3 ch s
We-should be inclined to attribute the expanding effect to
the influence of the attached atoms H, Cl, Br, etc., but the con¬
tracting effect to a residual action of the intense self-affinity
which characterizes carbon, and which persists in spite of the
presence of the other atoms.
Closely associated with the above influence is a third—
(c) The homogeneity or heterogeneity of the attached atoms .
In studying the methyl and ethyl compounds belonging to
the various series, we notice that when an atom or group other
than hydrogen takes the place of hydrogen, the atomic volumes
become smaller.
The amount of the diminution increases as the volume of the
heterogeneous atom increases.
The diagram on next page better illustrates this.
The two curves—that for the methyl and ethyl compounds
respectively, are very similar.
In both, we see that the volumes of the atoms diminish with
an increase in volume of the heterogeneous atom attached to
carbon.
Homogeneity then tends to expand the volumes, and hetero¬
geneity to contract them.
This is well shown in the following—
H
H
1
H
1
Cl
1
Cl
|
H—C—H
1
H—C—Cl
1
H—C—Cl
1
H—C—Cl
1
Cl—C—Cl
1
A
1
H
1
Cl
Ai
Cl
4*85
3*88
3*617
3*674
3704
Digitized by v^ooQle
LIQUID CHEMICAL COMPOUNDS.
Digitized by Google
ORGANIC COMPOUNDS CONTAINING OXYGEN 125
and also in the corresponding bromine compounds—
H H H Br Br
—H H—C—Br H— C—Bt H—i—Br Br—i-
H-
A
Br
V/W 4-85
H
384
A, L
3'895
3 ’ 7®4
A,
probably higher.
It will be noticed that there is an increase in complexity as
well as a change in homogeneity as we proceed from left to right.
It is difficult to consider them apart . The above values find their
place on a curve similar in character to the curves for the
normal series. The homologous increments in these cases are
—[H] + [Cl] and —[H] + [Br].
The effect due to homogeneity is much greater when the
volatile hydrogen is present, than when replaced by the heavy
chlorine atom.
In compounds like
Cl
Cl—Sn—Cl
b
Br
Br—Hg—Br < lb—Br
Ir
the halogen atoms possess their maximum values.
Exception.
Br
<
u
ir
Cl aa 22*2 Br 28*0 or over.
In this case, the atomic volumes of Br are
27*0, similar in fact to that of P, instead of 28 or
more.
Heterogeneity on the other hand usually results in a diminu¬
tion of the volume.
This has been shown to be the case in
CHgClg CHClj, CH s Br a
Cl
Br—L—Cl
A
Cl
NO a —( 1 —Cl Vm 110*9
ii
V m 108*4 CCIj 81*4 Br 27*0
CO,, 81*4 NO a 29*5
Normal 32*0
Digitized by v^ooQle
126
LIQUID CHEMICAL COMPOUNDS .
Thus the effect of homogeneity is an apparent repulsion
between the atoms attached to the central atom, and the effect of
heterogeneity is an attraction. These influences respectively pro¬
duce expansions and contractions.
The following cases are also interesting:—
Cl
0 = 1 *—Br V w = 107*4
P 0 C 1 , 79-4
|
Br 28*0
Cl
S»Va 107*4
Vm 107*4
which corresponds to
Cl H #
Cl—d—d Vm = 1067
i 11
Cl 0
CC 1 , 81*1
CHO 25*9
2 nVa 107*0
Vm 1067
The atom : O does not seem to be able to act in a manner
similar to that of the monatomic elements with single linkings or
with groups like N 0 2 singly linked.
It should be remarked that whereas chlorine atoms attached
to the same carbon atom give rise to expansions , or, perhaps, more
accurately, allow the volumes of the atoms to attain their full
value, on the contrary, when they are attached to different
carbon atoms, they cause contractions . This is an additional
feature which has already been noticed.
(i) Symmetry . The effect of this has also already been in¬
vestigated, at any rate in the case of the ethers.
Thus in the compounds C # H 14 0 and C 6 H 12 0 we find the
following characteristics.
H H
H H H
„-U- O— LU — H
Li LLi
v/w 3-659
H
H H H H
H— d —o—c—c—c— i: —h
i mi
3-643
H H
I
H
H H H
H. —*.** . ... •••!..—1 —C -
LLL LLi
-H H-
V/W
3*689
H H
.u^
u
H H H H
_u
u
3*666
-U
H H
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 127
In both of the above instances the complexity is the same,
as also the degree of compactness of the molecule, but the degree
of symmetry or distribution of carbon atoms about the oxygen
atom is different.
The most symmetrical compound possesses the largest volume,
and the values of V/W diminish as the degree of symmetry
becomes less.
In the ethers at any rate, degree of complexity modifies the
effect of symmetry, or rather want of it. Thus the effect of want
of complete symmetry among the higher compounds is not nearly
so marked as among compounds lower down in the series.
The effect of degree of symmetry on the volumes of the
atoms, and thus on the molecular volumes, is best shown by
some of the esters of the monocarboxylic acids.
TABLE LVI.— Isomeric Esters.
C 7 H 14 O a
M.V.
B.P.
C,H„
. CO
. 0
. CH,
172*2
149*6
C 4 H,.
,CO.
0.
C,H,
144*6
c # h 7 ,
. CO .
0.
c,h 7
174*0
i 43 *o
C t H,.
. CO .
0 .
c 4 h.
146*0
CH,.
CO .
0.
CjHjj
148*0
H.CO . O
.c (
.H u
173*3
153*2
C 8 H w 0 2 M.V. B.P.
C 6 H X8 . CO . O . CH 8 196*2 173*0
C 5 H n . CO . O . C 2 H 5 197*7 167*0
C 4 H 9 . CO . O . C 8 H 7 197*8 167*0
C 8 H 7 . CO . O . C 4 H 9 197*8 165*7
C,H 6 . CO . O . C 5 H n
CH 8 . CO . O . C 6 H 13 197*7 i6 9‘°
H . CO . O . C 7 H 15 196*7 176*0
We see that the most symmetrical compound possesses the
largest volume and the lowest boiling-point, whilst, on the other
hand, the least symmetrical compound possesses the smallest
volume and the highest boiling-point.
In the above series, as distinct from that of the ethers, the
least symmetrical compounds are at the beginning and ends of
the series, and the most symmetrical compounds are situated
near the middle .
In molecular volumes there is thus a rule analogous to that
stated by Hinrichs, 42 1868, for the boiling-point, which is, that the
more symmetrical the formula of an isomeric molecule is the lower is
the boiling-point This is similar to Naumann’s statement of it
(1874). The effect on volume is of the nature of an expansion.
The Phenolic and Other Ethers.
The group —O—R can be attached to a radicle of the type
C # H 5 - as well as to an aliphatic radicle R'.
R'—O—R C 8 H b —O—R
Aliphatic ether. Phenolic ether.
Digitized by v^ooQle
128
LIQUID CHEMICAL COMPOUNDS ,.
The phenols and similar compounds have already been
studied, and it has been shown that there is a contraction of 0*8
as compared with the alcohols, when the methyl group is present.
TABLE LVII.— Phenolic Ethers . 48
V w .
Phenol methyl ether anisol
Toluene
C 6 H 8 . O . CH,
C e H 5 . CH,
125*5)
Il8*2j
O'
7*3
Phenetol
Ethyl benzene
C 8 H 8 . O . C,H„
C c H 6 . c,h.
149-31
139-3 J
t °'
io-o
Phenyl propyl oxide
Propyl benzene
c„h 8 .0. c,h 7
c.h 8 . C 3 H 7
172-41
162-2 J
t °'
io-o
Phenyl butyl oxide
Butyl benzene
c 8 h 8 .0. c 4 h,
c 8 h 5 . c 4 h.
195*7)
184-5 j
0 '
11*2
TABLE LVIII.— Cresylic Ethers.
V*.
p Cresyl methyl oxide
p Xylene
C,H 4 (CH 8 ). 0. CH,
C„H 4 (CH,),
147-61
140-51
O'
7*i
p Cresyl ethyl oxide
p Ethyl toluene
C,H 4 (CH,) . 0 . c,h 8
C«H 4 (CH,). CjH,
H W
ff'S
oJ Ur
O'
10*2
p Cresyl propyl oxide
p Propyl toluene
C 8 H 4 (CH,) . 0 . c,h 7
C„H 4 (CH s ) . c,h 7
196-51
184-9 j
O'
ii*6
In comparing the
aliphatic and aromatic ethers,
in
respect of
the volume of O', we notice that it is only the methyl compounds
which differ, 9"6 and 7 2 respectively. When the groups are more
complex, there is very little difference between the volumes of O'.
A Study of the More Complex Aromatic Ethers.
The more complex ethers possess additional expansions owing
to their increase in complexity.
Thus
Phenyl heptyl oxide C a H„—O—C 7 H 16 V m = 271-4
C,H 6 928
O" io*o
C,H„ 158-9
2»V a 261-7
Observed 271-4
A = + 97 3 * 3-2 (Cr-Ci)
Phenyl octyl oxide C 8 H^—O—C 8 H 18 Vm = 296*8
C 6 H s 92-8
O" io*o
C 8 H 17 i8ro
2«V« 283-8
Observed 296-3
A = + 12-5 4 x 3-15 (C 8 —CJ
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 129
A slight increase is noticed at —O—C 4 H 9 , but after this there
is an increase of 3*2 for every addition of CH 2 .
The Thymols,
Methyl thymol C 8 H 7 . C 6 H 8 . (CH 8 )OCH 8 Vm = 214-8 *
Cymene C 8 H 7 . C 6 H 4 . CH S 184-7
less H 3-2
181-5
O' 7-4
CH, 25-5
214-4
V m 214*8
Ethyl thymol C 8 H 7 . C 6 H 8 (CH 3 )OC*H 5 V m = 240-5
Cymene C 10 H 14 less H 181-5
CjH b 48*0
O' ii-o
2nV a 2405
V m 240-5
C 6 H 4 89-6
O' 5-6
C 8 H 7 70-0
CH 8 255
5nVa 190-7
V m 189-3
-1-4 for 0 structure.
This is all the more remarkable seeing that the ortho cresylic ethers show it.
It follows from the above calculations that
—O—CH S gives a value of O' equal to 7-4
ZoUr* 1 * 6 an< * } 8* ves values of O' equal to 10-n.
When the total value of the side chains equals C 4 or C 6 there
is a large increase in the volumes. This is due to complexity
and involves every atom in the compound.
Since the volume of—OCH 3 is the same in aliphatic methyl
esters, and in aromatic methyl ethers, we are inclined to connect
the contraction with the presence of unsaturation. Note the
following:—
♦None of the thymol ethers show any contraction for ortho structure (vide
infra). Compare thymol itself (q.v.)
9
Digitized by v^ooQle
130
LIQUID CHEMICAL COMPOUNDS.
o=c—
Jc
iCH a
c=c<
OCHg
/OCHg
— CH'
\0CHg
in all of which >0 = 7—8, and compare with this R—O—CH 3
in which >0 = 9*6. The conclusion before arrived at, seems the
only one possible.
The Cresylic Ethers.
p Cresyl butyl oxide C 6 H 4 (CH S )0 . C 4 H 9 Vm = 221*3
C
-CH,
c 6 h 4
8g*6
/ '
\
CHg
25*5
CH
CH
O"
10*0
|
||
C 4 H 9
92*3
CH
CH
2«Va
Vm
\ .
C
/
—0—C 4 H„
217*4
221*3
A = + 3*9 ix 3*9(C 6 -C 4 )
p Cresyl octyl oxide C 6 H 4 (CH 3 ). OC 8 H 17 V m = 323*1
C-CH,
c,h 4
8g*6
s \
CH,
25*5
CH CH
1 II
0"
10*0
C 8 H„
181*0
CH CH -
\ /
2»Va
306*1
C—0—C.H,,
Vm
323*1
A = + 17*0 5 x 3*4 (C 9 - C 4 )
The Thymols.
Propyl thymol C 3 H 7 . C 6 H 3 (CH 3 )OCgH 7 V m 267*5
CH
CH,/
CH—CH
*\,CH—C^ ^C—CH,
Cio^is 181*0
™ 10*0
\
=CH
/
&
c 3 h,
70*0
o-c 9 h 7
2nVa 26l*0
Vm 267*5
A = + 6*5 2 X 3*2 (C 7 —C B )
Heptyl thymol C 8 H 7 . C 6 H 3 (CH 3 )OC 7 H 15 V w = 369*5
^10^13 iSi'O
CH—CH
CH *\ / \
\CH—C x C—CHg
ch,/ \ /
C—CH
I
o-c 7 H ]5
O' 10*0
C,H J5 i 59 'o
2 »V* 350-0
Vm 369-5
A = + 19-5 6 X 3-2 (C u - C 6 )
Digitized by V^.OO £
ORGANIC COMPOUNDS CONTAINING OXYGEN. 131
Octyl thymol C,H, . C,H,(CH,)OC 8 H 17 V* = 396-5
CH—CH C «* H i? 181-0
CH v O io*o
>CH —or ^C—CH, C 8 H 17 182-8
CH,/ \ / -
C=CH 2 »Va 374*8
I V« 396*5
6—c 8 h 17
A = + 21*7 7 x 3*i (C 13 -C B )
The increase in volume for complexity is seen to be appar¬
ently linear. This increase starts from approximately the same
point in all the series, viz. where the total complexity of the side
chains is C 4 . Apparently it does not matter whether the chains
are entire , or divided up , for this to occur . This indicates an
increase in volume due to a corresponding increase in com¬
plexity as distinct from an increase in length of chain. The
increase starts from C 5 which is similar to what is observed in
open chains. We conclude that the nucleus exerts no influence
at all. Octyl thymol is interesting, as being the most complex
compound which has been studied in this work. It consequently
possesses the largest volume which is one of nearly 400 units.
The calculation of the volume of an ether of the type men¬
tioned is as follows :—
p Cresyl heptyl oxide.
CH,—C,H 4 —
-OC 7 H 15
C.H.
8g*6
CH,
25*5
O'
11*0
c 7 h 16
158-9
285-0
c,—c 4
+ 12-8 (4x3-2)
SnVa
297-8
Vm
298-3
The o y my and p Cresylic Ethers .
It has been shown that when two groups occupy the
ortho-position, there are usually contractions.
C 8 H 4 =8g*6 CH 8 =25'5 C 2 H 5 =48 0 '= 7*4 or iro C»—C 4 =3*2 (#—4).
o m p
^ CH
CH
cu
CH
/
— ch 3 — 1 CH 3
CH CH CH OH
CH C—O—R CH CII
^CH^ ^qZ-O—R
9 *
Digitized by v^ooQle
132
LIQUID CHEMICAL COMPOUNDS.
Groups CH 3 and OCH s
146-4
I47’8
148*0
2nVa
148*0
A -1*6
-0-2
± 0
CH„ OC a H 5
171-2
1727
172*8
173*1
-1-6
-0*1
+ 0
CH,. OC 8 H 7
195*4
1967
196-5
195*1
- i-i
• —
± 0
ch 8 , oc 4 h 9
218-8
221*0
221*3
221*4
-25
-0-3
± O
CH 8 , OC 7 H 1b
2936
297*4
298-3
298*0
-47
-0*9
+ O
CH„ OC 8 H„
3I8-6
3227
323*1
323*4
-4*5
-0*4
± O
The magnitude of the contraction evidently increases as
—OR becomes more complex, thus showing the existence of a
special relation between the two groups.
There is thus a large contraction when the groups are in the
ortho-position relative to each other. This appears to increase in
magnitude as the complexity of R increases. The above con¬
tractions are regarded as being due to a mutual action of the
group, owing to residual affinity.
The following compound is worthy of separate study:—
Dimethyl resorcin.
157*7
C-och 8
c 6 h 4
89*6
\
CH CH
20'
14*8
1 H
C C—OCH 8
2CH 8
52*o
\ /
CH
2nV a
156-4
v m
157-7
The oxygen atoms are probably augmented in value, as com¬
pared with its value in previous compounds, owing to the fact
that there are two—8*0 e.g. The value of O' in - OCH s is,
however, still considerably below that for O . C 2 H 6 - io units.
It will have been repeatedly observed that the methyl group
always causes a diminution in volume as compared with the
ethyl, etc.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. 133
Oxygen with Double Linking.
= o.
(Attached to carbon.)
The simplest compound is—
Carbon monoxide < C = O V wt 22*1.
Carbon in this instance is divalent, and it has not been shown
whether divalent carbon possesses the same volume as quadri¬
valent carbon or not An examination of such groups as CHO
and CO, under certain circumstances , shows that in combination,
CO = 22*1.
C 8 H 7 . CHO 96 0
^ 3^8 74 *°
CO 22*0
It follows that the oxygen under these circumstances (i.e.
when carbon is quadrivalent C iv ) is the same as when it is
divalent.
O" = (CO - C) = 221 - 14-8 = 7 3 (H a )
Carbon dioxide C [( V« «*n
V« 33*0
The volume of C = 14*8 .*. O a = 187.
It is remarkable that the volume occupied by 0 2 is the same
as the volume of 0 2 in the esters - Cn
and similar to the
as cue vuiumc ui w 2 in uic cslcis — V '\0(R) a,Ilu Slllllliu LU U1C
one calculated from the acids - C^5 /t rv.
Moreover if the volumes of certain compounds containing the
group - C^q are considered, these compounds being derivatives
of the acids, we see that the volume of : O is 11 *o.
CHo- Cr r»" _ ^ _l oo. T \ _
O" = 74-(25-9 + 14*8 + 22*1) = 11*2.
Moreover [O"] = 33*5 - 22-1 == 11*4.
It is possible that the two oxygen atoms may possess
different volumes (a) [O'] = 7*4(2!!), (b) [O''] = ii o(3H).
[O"] . 11*4 )
- Cf [ i8*8 for 0 2
\0H [O'] = 7*4 J
- Cf [O"] = (48-0 - 14-8 - 22*1) = iro.
\C 1
Digitized by LjOOQle
134
LIQUID CHEMICAL COMPOUNDS.
An alternative supposition is that the volumes of oxygen
(O'') are similar, i.e. [O"] = i8‘8 -s- 2 = 9*4.
The Aldehydes and Ketones.
In dealing with compounds which contain doubly bound
oxygen, we come to the conclusion that the position of this atom
in the side chain has some influence on the volume. Comparing
the ketones and the aldehydes we find:—
CH 8 . CH 2 . CH CH g —C—CH 3
II II
o o
Acetaldehyde Vm = 74*8
CH 3 . CH 2 . CH 2 . CH
. II
O
Acetone Vm = 77*1
CH 3 . CH a . C . CH 3
II
O
Methyl ethyl ketone V« = 96*7
CH 3 . CH a . CH 2 . C . CH S
II
O
Methyl propyl ketone Vm = 118*5
CH 3 . CH 2 . C . CH a . CH 3
II
O
Diethyl ketone Vm = 117*7.
These compounds are difficult to deal with.
(a) On the one hand, we have one factor involved, which is
the position the compound occupies in its particular series.
We should expect the acetaldehyde and acetone to be rela¬
tively larger than the volumes calculated on the basis of C=* 14*8,
H = 37, and thus be not quite comparable with the succeeding
members of each series.
(b) There is also the question of the influence of symmetry-
Both of these factors might be considered to be of influence on
the volumes, and to explain the differences.
Observation, however, shows that there is something wanting
in this kind of explanation.
The study of molecular volumes has made the want of a com¬
prehensive generalization, which would cover all these cases, very
keenly felt. The good results which have attended the study of
some series show that we are getting nearer to the discovery of
such a generalization.
If we relinquish such purely geometrical ideas as symmetry,
and introduce the more chemical notion of a specific influence of
Propaldehyde Vm = 96*0
CH 3 .CH 2 .CH 2 .CH 2 CH
II
O
Butyraldehyde Vm = 118*2
Digitized by CnOOQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 135
certain atoms or groups, we seem to be getting nearer to the
desired end. After all, the question of symmetry may be in¬
volved, but only in a subordinate sense. The particular arrange¬
ments of the atoms in a compound are bound to have some
influence on the volume, and among them their symmetrical
arrangement about the distinctive atom or group of the compound.
It is possible for there to be a yet more fundamental explanation,
which may lie in the direction of the alternative one already
suggested.
It must especially be noted that the volume variations in the
series of compounds including the aldehydes and ketones of
various kinds seem inseparably connected with similar variations
in the boiling-points. Compounds with: O at the end of the chain
have the smallest volumes and boiling-points. The volumes ap¬
parently increase as : O becomes attached to intermediate carbon
atoms.
It has been shown that when the iso group is present in a
pure hydrocarbon (paraffin, etc.), there is a contraction.
ch 3 —ch—ch 2 —ch 3
r a = — o-s
I'M,
CH,
CH 3 —C—CH a —CHg A = - 2 x 0*5 = - ro
These contractions are connected with the presence of methyl
groups which probably are not saturated. In such a modification
as—
CH 3 —C=CH—CH S
L a = " 2 ’ 4
there is an increase in the contraction, owing to the presence of
the. ethenoid linkage, in close proximity to the iso group. In
some paraffin derivatives we, however, get a different result:—
CH,—CH—Cl
in,
CH a — CH—CH„—Cl
U
A = +2*6
A = + 1*5
Whilst the association of two hydrocarbon radicles, or a
hydrocarbon radicle and one which contains an ethenoid linkage,
Digitized by v^ooQle
136
LIQUID CHEMICAL COMPOUNDS .
occasions a contraction, the association of a hydrocarbon group
and a halogen atom produces an expansion. It is evident that
the methyl CH 3 group exerts a more powerful influence than
an ethyl group, and the extent of the influence of the halogen
atoms is in the order indicated by chlorine, bromine, and iodine.
In studying the groups, we find that there is a depressing
effect due to the simple radicle, which diminishes as this becomes
larger. The effect is thus in the inverse order of—
H, CHg—, C a H s —,*C a H 7 —.
We have now to consider the influence of H, CH 8 , C 2 H 6 , etc.,
on C : O. As in the halogen compounds, the substitution of: O
for H 2 increases the volume. The atom hydrogen does not seem
to have much effect, but that of the methyl group is considerable,
whilst it is progressively smaller for C 2 H 6 and C 8 H 7 —
/H
(1) ch, . cr
'' \o
Vm= 567
2nV a
51*9
A + 4*8
(CH S and H)
/h
(2,C ’ H ‘<o
t
II
60
>»
74 *°
+ o*8
(C s H 5 and H)
/CH,
( 3 ) CH, . /
\o
V w = 77*1
••
74 *o
+ 3*1
(two CHg)
/H
4 ) c,h 7 . /
\o
II
%
6
»»
96*0
—
(C 3 H 7 and H)
/CH,
( 5 ) C,H,.C(
\o
V m = 967
>»
96*0
A + 07
(CH 8 and C 2 H 5 )
/H
(6) c 4 h,.c;
\o
V m = n8*2
Il 8’2
A —
(C 4 H 9 and H)
/CH,
(7) c,h 7 .c(
\o
/C,H,
(8) C,H,. c;
\o
V w = 118*5
»>
118*2
A + 0*5
(C 8 H 7 and CHg)
V w = 1177
>»
Il 8*2
>
1
0
ut
(two C 2 H b )
It is seen that wherever we find one
or two
methyl groups
we find a large positive effect. That due to the C 2 H 6 group is
less.
Whilst the above seems clear, we must also consider another
factor—the total complexity of the compound. The effect of the
methyl group diminishes as the total complexity increases. In
number (8) the combined effect of complexity and the C 2 H 5
group is to produce a negative difference, but in number (7) the
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. 137
expanding influence of the CH 3 group is sufficiently great to
convert this into a positive difference. It is extremely probable
that were complexity alone operative, the relative volumes would
steadily diminish to a minimum at the first member of the series.
Seeing, however, that the first members in each series include
the methyl and ethyl groups, there is a contrary or expanding
influence which is a maximum at the first member of each series,
and this more than balances the effect of diminishing complexity.
The Aldehydes.
It is evident, as Kopp himself discovered, that the aldehydes
possess similar volumes to those of the paraffins, so that : O may
be supposed to occupy as much space as two volumes of
hydrogen.
TABLE LIX.
The Aldehydes.
Acetaldehyde CH S . CHO
Propaldehyde CH S . CH S . CHO
Butyraldehyde C 8 H 7 .CHO
Valersddehyde C 4 H 9 . CHO
Benzaldehyde C 6 H B .CHO
v m .
v M .
The Paraffins.
567
—
C a H,
74-8
—.
c»h 8
96*0
96*0
c 4 h 10
Il8*2
117*8
C»H ia
118*4
118*2
C.Hj. CH S
[s 0] = 7 ’ 4 .
The two series are strikingly similar, so that we may suppose
that the volumes of the simpler paraffins resemble those of the
aldehydes.
Acetyl chloride CH S . COC 1
H
H—C—C .
A i
Vm 74*1
CH S 26 0
C 148
Cl 22*2
O" iro
SnVd = 74*0
Wm = 74*1
Chloral C^HO. Vm 106*37.
Cl—c—c = 0
V CCI3H
84*5
n
less H
-37
80 *8
CHO
25*9
2nV a
106*7
Vm
106*37
Digitized by v^ooQle
LIQUID CHEMICAL COMPOUNDS ,,
138
Trichloracetyl chloride C 8 C 1 4 0 . V», = 125*5.
Cl
1
V CC1 3
c
0"
8o*8
14*8
ii*i as in - C0"C1
Cl-C —
1
- 0
II
0
Cl
22*2
1
1
2 »v a
128*9
. Cl
Cl
less for ai 8 subst.
-3* 1
2»Va 125*8
V w 125*5
The Ketones.
Acetone CH 3 —CO—CH S Vw 77 * 1 *
The volume is considerably larger than that of the isomeric
propaldehyde, from which we might conclude that the volume is
[:0] = 74 + 23 = 97-
The following methyl ketones, except the third and fourth
which are interpolated, have been investigated by Thorpe.
TABLE LX.
The Methyl Ketones . 44
Compound.
C 2 H 5 . CO . CH S
C 3 H 7 . CO . CH S
C 4 H 9 . CO . CH 3
c 5 h u . CO . CH 3
c 6 h 13 . CO . CH 3
v„.
v w .
The Paraffins.
967
96*0
c 4 h»
118-3
117*8
C 5 H i2
;i 4 o-i)
139*9
C.H 14
162-8)
162*6
C,H 1#
186-6 (Sch.)
186*3
c 8 h I8
The volumes of these methyl ketones are a little larger than
those of the corresponding paraffins.
Pinacolin
ch 8 )
CH. J-C . CO . CH„
ch 3
c.h 12 o.
less for
Vf» 138*5.
^6^12 I 3 2 *4
O 7*4
139*8
^group - i*2
- 138*6
SnVa for pinacolin C tf H 12 0 2 (139*8) is not far from that of
C 6 H 14 nor from that for C 4 H 9 . CO . CH 3 the normal compound.
The value of the group is about equal to two iso groups.
Since CH 3 . COC 1 represents the acetyl group + a chlorine
atom we see that : O may possess a volume larger than 7*4, one
in fact equal or nearly equal to iro.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. 139
An interesting case is that of
Acetoacetic ester C 6 H 10 O 3 .
V m 153*6
CH 3 —C—CH 2 —
11 \oc a H 5
0
The compound has
V w 150*5
CH 3 —C—CHj-C^ 0
II \OCjHj
A + 3*1
If the constitution of acetoacetic ester were “ enolic ”
CH3-c • CH
OH
C
^ OC2H5
the volume would be similar to that of ethyl butyrate, or even
less as the matter stands, the constitution cannot be other than
“ ketonic *\
This is interesting from a chemical stand-point, and confirms
the conclusions derived from other sources, both chemical and
physico-chemical.
There is one point which may have an important bearing
upon unexplained variations in atomic volumes from the normal
when we form a more or less complex substance by the union or
condensation of two simpler ones, one particular atom of which
has a definite value in one compound, and remains unaltered in
the complex one. Thus in the synthesis of aceto-acetic ester:—
ch -<
H
CH S . COO Et
CH S . C-CH a • COO Et
II
o
The atom O" is thus similar to carboxylic oxygen in ethyl
acetate, and is also similar to the atom in:—
CH -<C,
The group—CH S CO is really an acetyl group and O" only
apparently resembles a ketonic oxygen.
Digitized by v^ooQle
140
LIQUID CHEMICAL COMPOUNDS.
One other compound remains, also interesting, from a chemi¬
cal point of view, viz. Camphor
Camphor C 10 H 16 O. V m 1877.
CH S -CH-ch 2 We can obtain its value from that of
borneol Vm 190*5.
Group OH = 6*o •+■ 3.7 = 97
c io H i5 = I 9°*5 - 9*7 = 180*8
less H 37
57 x 37
less for ring
2 tNa
Obs.
0"
2 nV a
218*3
- 30*0
177*1
11*0
188*1
188*3
187*7
The Carboxylic Acids . 45
The aldehydes and ketones give rise to the
and
(a) R “ C \ oh and W R — 1 Cf
\0R
Carboxylic acids. Carboxylic esters.
These compounds contain the oxygen atom * O •
doubly-bound one : O.
and the
TABLE LXI.
Substances.
Vm.
»CH S .
A for O x .
Propionic
. . . C,H,O a
85-3
66-3
19*0
Butyric.
. . . C 4 H 8 O s
107*8
00
00
19*4
Valerianic
. . . C fi H 10 O 8
130*0
iio*5
19*5
Caproic.
• • • CjHijOj
152*5
132*6
19*9
Heptoic
. . . C 7 H 14 O a
174*6
iss^
19*4
Mean value =
19*5
These acids are altogether exceptional, in that, although they
contain the hydroxyl group, their volumes are by no means
diminished in consequence.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 141
The value of: O may be obtained as follows :—
TABLE LXII.
The Fatty Acids.
V m .
0 ".
v w .
The Paraffins.
C.H.O, . .
853
n *3
74*0
C,H S
c 4 h 8 o, . .
107*8
n*8
96*0
c 4 h 18
C,H 10 O, . .
130*0
12*2
117*8
C.H„
C 6 HiA
152*5
12*6
139*9
c«h 14
C 7 Hi 4 0 2
174-6
12*0
162*6
c 7 h„
The average value of O” is 12*0, the assumption being that
[*0*] = 2 H
for [ 0 "] + [O'] = 12 x 7*4 = 19-4-
The value of O" can also be obtained from the compounds
containing the group - C\Q*
Phosgene Cl - C!
\ m 7 °'°
/°
\ci
O" = 74*0 - (148 + 2 x 22*1) =* 11*0
Acetyl chloride CH, - C:
-c/°
\ci
M.V. 74*0
Q^' = 74*0 - (14*8 + 25*9 + 22*1) = ii *3
From acetic anhydride we obtain the value of O'
CH 8 . CO . O . CO . CH 8
2 Vols. CHj,. COOH
CH S . CO - O . CO . CH 8
M.V. 109*9
125*6
109*9
H a 0 157
H 2 7*4
•O* 8*3
O" + O' = n*3 + 8*3 = 19 * 6 .
The excess is thus thrown on to the O" or on to the O'
according to the method of calculation.
The simplest explanation is to suppose that [ 0 /7 ] = 3[H]
and [O'] = 2[H].
Then, assuming that 0 2 = 19 5
[O"] =* 11*7 and [O'] = 7*8.
Digitized by v^ooQle
142
LIQUID CHEMICAL COMPOUNDS.
The Methyl Esters . 46
The methyl esters are sharply distinguished as regards the
volumes of oxygen from the acids on the one hand, and from the
more symmetrical esters on the other.
TABLE LXIII.
Radicles.
The Acids .
The Methyl
Esters .
Symmetrical .
Vm.
A.
v w .
A.
v w .
CH S —CO—
853
- 2*1
83*2
+ 1*4
84*6
H . C0 2 C 2 H 5
C 2 H 5 —CO—
107*8
-3*2
104*6
+ i*5
106*1
CHg . C0 2 . C 2 H 6
CgH ? — CO—
130*0
-3*3
1267
+ 17
128*4
CH, . C0 2 . C 3 H,
C 4 H 9 —CO—
I 5 2 '5
“3*4
149*1
+ i*4
150*5
C 3 H, , COj * CjHg
C 5 H u —CO—
174-6
“2*4
172*2
+ 1*8
174*0
C s H 7 . CO a . C,H,
Mean =
-2*9
Mean =
+ 1-6
We find that by comparing—
and also
- C \OR
and —Cf
\OCH,
A -2*9
and
A +i*6.
_C^°
\oCH,
This result is quite different from that shown by the alcohols
CH 2 . OH, as compared with CH 2 O . CH 8 and CH 2 —O—R.
The alcohols are depressed, whilst the acids are augmented in
volume.
As regards the ethers and esters, the O—CH 8 group produces
a diminution in volume in both.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. 143
TABLE LXIV.
Monocarboxylic Esters.
(Second Series.)
Substances.
v w .
»CH 2 (22*l).
o 2 .
HC 0 . 2 CH 3 ....
62*6
44*2
18*4
HC 0 2 C 2 H 5 ....
84-8
66*3
* 8-5
H . COjCjH, ....
106*4
88*4
18*0
CH,. CO a C,H 7 . . .
128*8
110*5
i 8*3
c 2 h 5 . co 2 c 8 h 7
151*0
132*6
18*4
Mean value =
18*3
TABLE LXV.
The Esters.
C#Hjj«O a .
v w .
0".
v m .
The Paraffins.
C»Han + 2*
HCO a CH 3 .
62*6
11*0
51*6
CjH 8
hco 2 . C a H 6 .
84*8
10*8
74-0
C s H 8
h.co 2 .c 3 h 7 .
106*4
10*4
96*0
C.H W
CHg . CO a . C 3 H 7
128*8
11*0
117*8
C 5 H ia
c 2 h 5 . C 0 2 . C 3 H 7
151*0
ii*i
139*9
C.H„
Mean = 10*9
Thus assuming that [. O'.] = 7*4 (2H) [O"] = 10*9
[O"] + [O'] - 7*4 + io *9 - i 8 * 3 .
We thus have—
for the acids
O,
19*5
O"
117
O'
7-8
the methyl esters
o a
16*6
0"
11*0
0'
56
and the symmetrical esters .
o a
18*3
0"
11*0
0'
7‘4
There is distinct evidence that [O"] = 3H and [O'] = 2H
throughout the series, in the following:—
Digitized by v^ooQle
144
LIQUID CHEMICAL COMPOUNDS.
TABLE LXVI.
Symmetrical.
Esters.
V m .
Vm.
Ethers.
CnHjnO.
C a H 4 O a
62*6
62*5
Cn H a n-f- a O
c 8 h 6 o
C.H.O,
c 4 h 8 o.
84*8
106*4
106*4
c,h 8 o
c 4 h 10 o
C 6 H 10 O a
128*8
128*0
C,H w O
C 6 HiA
151*0
i5i*3
c,h 14 o
c,h 14 o.
i 74-6
174*7
C 7 H t i,0
^8^16^2
*97*7
197*7
c 8 h 18 0
<^° d
^10* 1 30 U 2
221*9
221*1
c»h m o
245*9
246*1
C 10 H ffl O
CnH M O a
270*3
271*3
C n H M 0
^12^94^2
295*9
295*5
C u H„0
^14^28^2
35i*2
353*4
C 14 H m O
^15^80^2
377*0
377‘6
Cj.H.,0
4°4*3
404-5
c 16 h m o
Thus, in spite of the change due to complexity
[O'] - 2[H] and [ 0 "] = o 0 ] = 3 [H]
under ordinary conditions [O'] = 7*4 [O"] = [> O] = 11*0. In
both series we find contractions when the group - OCH 3 is
present (a) A diminution of about — i -o in the ethers and ( b )
a diminution of - i*8 in the esters. This may be due to a
contraction of the O' atom alone, of the whole group, or of only
the CH 3 group under the influence of the oxygen atom.
The Effect of the Homologous Increment CH 2 on the
Volumes of the Monocarboxylic Esters.
These compounds have been studied chiefly by Gartenmeister
(loc. cit.), Schiff, and Elsasser.
The series of the methyl esters may be made use of to illus¬
trate Lossen’s method of calculating the volumes of series of
compounds.
Thus, the volume of methyl formate HCOOCH 8 is V m 627,
and the difference between this volume and that of methyl
acetate CH 3 . COOCH 3 is A = 20*5.
The assumption is made that the volume of the homologous
increment CH 2 increases by 0*5 from term to term.
Thus for any compound C»H 2 » 0 2
Vm = 627 + (» - 2) 20*5 + 0*5 ~ 2 ) 8 (1)
2
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 145
Now Kopp had made the assumption that
[HJ = [O], (for both O' and O")
and this, combined with the assumption that the average initial
difference for CH 2 , is
A CH a = 20*9
enabled Lossen to calculate from methyl formate
C 2 H 4 0 2 . V m =b 627
the values [C] = 10*45 [H] = 5*225 [O] = 10*45
Then the above equation becomes
V#» = I0*45» + 5*225W + 10*450 + o*25(» - 2) 2 ,
and by its aid, the volumes of the different members of the series
can be calculated.
A different method is adopted here, which is based on the
idea that whatever may be the changes in volume in a homol¬
ogous series, the relative volumes remain the same throughout.
If we assume that
c = 4 h, O' = 2H, O" = 3 H
in the formic esters, we may obtain a curve which shows how the
atomic volumes change throughout the homologous series. This
is accomplished by finding the sum of the hydrogen equivalents
in the different compounds, and the quotient of the volume
V m and this sum W, gives the volume of hydrogen in a particular
compound. By this means we avoid the necessity of making
use of the difference in volume between two consecutive members
of a series, which is neither true for the one compound nor the
other.
Thus for compound C»H 2W 0 2
Vm = (6» + 5)S = WS.
TABLE LXVII.— Volumes of the Atoms in the Formic Ester Series.
CmH 2 » 0 2 .
Compound.
w.
Vm.
Vm
W*
Methyl formate
. hco 2 ch 8
17
627
3*688
Ethyl „
. HCO s C 2 H fi
23
84*6
3*678
Propyl „
. HC 0 2 CoH 7
29
106*2
3*666
Butyl „
. HCO 2 C 4 H 0
35
127*6
3*645
Amyl
4 i
150-5
3*671
Hexyl „
. hco 2 c 6 h 18
47
173*3
3*688
Heptyl „
53
196*7
3 * 7 H
Octyl „
. HCO a C 8 H 17
50
220*3
3*734
IO
Digitized by v^ooQle
146
LIQUID CHEMICAL COMPOUNDS /
10 20 30 40 50 60
Number of H Equivalent's.
Fig. 6.
The above curve shows how the volumes of the atoms, and
thus of the group CH 2 , vary in the formic ester series.
The curve is very similar to the methyl ether curve, and
shows a well-marked region of diminishing values of V/W. The
region of the minimum is here reduced to a point, as the curve
is formed by the intersection of two others. The ascending
curve is as far as we can make out linear, with a value of
A V/W = 0*022 on the average. The curve as a whole, whilst
in many respects resembling that for the normal paraffins, is in
others quite different
These curves do not seem capable of exact mathematical
formulation, but this is not of great consequence, as the mole¬
cular volumes of compounds are calculated in quite a different
way. It is much better to consider the total variation from a
volume calculated from mean values, and regard this total
variation as due to general or particular constitutive influences.
An estimation of the values of these is often possible. The
curves are, however, useful in showing how the atomic volumes
vary in a series, and also in indicating points of similarity or
contrast with the curves for other series.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. 147
So far as the methyl ester series is concerned, we find that—
S' = 3*645 + (6 n ~+~5 - 35) x 0*0037
since 1/6 of 0*0022 is = 0*0037
and the total volume of a compound is
Vtn = ( 6 n + 5) {3*645 + (6 » - 30) x 0*0037}.
If we compare the volumes of certain of these compounds,
we get the following constants:—
H.COOCH3 v m 6270
H . COOC 6 H 18 Vtn 173*30
5CH a 110*6 CH a 22*12
Since H . COOCH 8 is C a H 4 O a
O a = 6270 - 44*24 = 18*46
NOW O" II ^} 18-5 for ° a CH » = 22-1 -
Although the results obtained by Lossen's method and the
one just described are both in good accord with the experimental
results, yet there are some points of difference between them.
The most important of these is the difference in the curves
representing the atomic volumes in different members of the
series.
Lossen assumes that the expansion is rectilinear from the
first compound, whereas the curve just indicated shows a con¬
traction to a minimum at the fourth member and then an ex¬
pansion. In this respect the methyl ester curve is similar to all
the other curves studied.
The Methyl, Ethyl, and Propyl Salts of the Fatty Acids.
This is an arrangement of the esters which is quite different
from the other. In the first, the acid is the same, but the alkyl
groups are different; in this arrangement the fatty acid radicles
are different, but the alkyl group is the same for all. It by no
means follows that the relations between the atomic volumes are
the same in the two methods of arrangement. It has been
found that better results and smoother curves are obtained by
utilizing Schiff’s data as well as those of Gartenmeister. The
methyl ester curve is wholly from Gartenmeister’s results, the
ethyl ester curve is partly due to Schiff* and partly to Garten¬
meister, whilst all the values employed in the formation of the
10 *
Digitized by v^ooQle
148 LIQUID CHEMICAL COMPOUNDS,
propyl curve are due to Gartenmeister except that for propyl
propionate.
We give below a diagram showing the volumes of the esters
of formic and acetic acids. There is a very great difference
between the first part of the two curves. On comparing the
formulae
H - COOR CH S - COOR
The formates. The acetates.
we see that this difference is to some extent connected with the
different effects of the two groups
H
and
The CH 3 group again is responsible for a contraction. It is, how¬
ever, difficult to analyse the data completely, because at this point
the effect and change due to complexity is enormous. The best
results are obtained if we can compare the volumes of different
compounds of similar complexity. Some of these effects are very
complex, and are the sum total of the combined effects of the
atoms and groups.
Number of H Equivalents
Fig. 7 .
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. 149
H
<
fa
W
X
H
J
<<
CO
>«
a,
O
%
a.
Q
2
<
S
>*
x
H
W
J
>«
X
H
M
s
>
X
w
CQ
<
H
3
13
CO
I
2
13
CO
3
w
fit
vo
8
CO
00
C*
O
pt
<0
8
C*
O'
vp
VO
VO
tv
tv
tv
tv
>
CO
eo
Jo
<0
Jo
Jo
Jo
O
0
00
N
to
*
00
H
tv
«
vb
>
<M
to
tv
O'
et
rt-
H
H
w
H
N
N
4
X
sf
CO
O
N
K
CJ
X
CO
CJ
e"
CO
0
af
d
E
eo
CJ
3
8 .
6
0
0
C*
O
CJ
O
cj
0"
0
d
0
0
CJ
0
CJ
O
CJ
0
CO
X
X
e
o»
E
B
•0
H
E
X
CJ
0
0
CJ
O
CD
O
l*
S'
to
CO
M
tv
rf-
co
«o
O'
to
to
VO
rs*
CO
00
VO
10
8
R."
1
1
VO
VO
VO
VO
1
1
1
>
Jo
CO
Jo
Jo
tv
H
H
to
V
vb
00
b
I
1
1
>
00
0
0*
to
1
1
1
H
H
H
X
a?
10
E
E
V)
E
10
E
4
10
X
oc
cj
0
0"
O
O
CJ
l
B
M
CJ
O
0
0
0”
0
< 3 *
O
O
O
0
O
CJ
eo
O
O
0
0
•
■0
r»
at
X
3
0
.
sT
e
SB
X
E
E
X
0
0
0"
0
df
dr
>
CO
O'
«o
M
tv
co
O'
01
M
CO
rt-
10
»o
00
s
tv
M
vp
N
8
T
VO
CO
vp
sf
0
tv
0
tv
>"
to
Jo
Jo
Jo
Jo
Jo
Jo
Jo
j
tv
«
op
Jo
VO
V
tv
vb
H
O'
«
«
«
vb
M
b
>
VO
00
0
rt-
tv
O'
et
M
H
H
H
M
N
eo
00
50
4
G
X
CO
X
CJ
CO
X
CJ
CO
E
0
co
E
O
E
cj
E
CJ
E
cj
i
1
CJ
CJ
09
O
CJ
cc
O
O
0
0
10
0
cj
Cl
O
CJ
A
Cl
0
0
0
CJ
CO
d
CJ
id
•
X
X
E
E
a?
X
E
X
CJ
0
CJ
0
CJ
0
d
£
N
CO
O'
»o
H
tv
CO
O'
H
N
N
co
rt-
rt-
to
to
In the above table, we notice the general effect of change
of complexity on the atomic volumes, and this is also indicated
by the curves. These curves are totally unlike those shown to
be applicable to the esters of the same acid. They, however,
resemble the curves for the normal paraffins and other com¬
pounds. (a) The descending portion is more pronounced for
Digitized by v^ooQle
150 LIQUID CHEMICAL COMPOUNDS .
the methyl salts than for the others, for the reason that the
simplest compounds of the series have shorter chains, and are in
general of smaller complexity. ( [iY) The curves are wide and
open, especially that for the methyl salts. This is totally unlike
the curve for the methyl ethers or for the esters of formic acid
which are broken at the minimum, owing to an abrupt change in
the curvature.
10 20 30 40 50 60 70
Number of H Equivalents.
Fig. 8 .
An Examination of the Causes of the Variation of the
Volumes of Isomers, and a Consideration of the General
Question of the Influence of Symmetry.
It was shown by Gartenmeister that the volumes of com¬
pounds possessing similar formulae, but differing in the method
of the distribution of the carbon atoms between the O and C
radicals, possess somewhat different volumes. Thus in a com¬
pound
R—C
\
o
OR'
R and R' may have different complexities, and the volumes of
the compounds depend on this circumstance to some extent.
Similarly the boiling-points vary.
The following series shows this.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 151
TABLE LXIX.— The Isomeric Esters of Formula C 8 H 16 O a .
Compound.
Vm.
b.p.
C 6 H 13 . COOCH3
C 5 H 12 . COOC 2 H 5
196*2
173*0
1977
167*0
C«H n . COOCjH,
C 8 H 7 . COOC 4 H 9
ch 8 . COOC 6 H 13
197-8
167*0
197-8
165-7
197-7
169*0
H . COOC 7 H 15
196-7
176*0
The volumes are seen to be a minimum at the extremities
of the series, and a maximum at the centre. The opposite is
true of the boiling-points.
This has been ascribed to differences in the symmetry of the
molecules. Compounds in which the radicles differ most, possess
the smallest volumes, and those in which they are most nearly
equal possess the largest volumes. The latter are the most sym¬
metrical compounds, and the former the least symmetrical. We
believe that a consideration of such influences as symmetry,
which are incapable of accurate measurement, is undesirable and
that it will be better to substitute considerations of the specific
influences of groups.
Influence of the C and 0 Groups .
In the above compounds the groups
- O . CH 8 and H . C =5
produce contractions of - I *6 and - I * 1 respectively. The former
group is a frequent cause of contraction in volume as will be seen
from Table LXVIII on the methyl and ethyl esters. The mean
difference between the volumes of methyl esters and those of the
corresponding ethyl and propyl compounds is - 1 -6, a number
which is exactly similar to the one just given. The volumes of
the methyl esters are thus below the normal, whilst it may be
shown that the symmetrical compounds are normal.
Thus for
HCO a C a H 5 Vw 84*6
C 2 H 5 48*0
CO a 33 3
H 37
SttVa 85*0
Vm 84*5
The volume of
ch 8 . co 2 . CH S Vm = 83*2
is thus 1 *4 below the normal,
Digitized by v^ooQle
152 LIQUID CHEMICAL COMPOUNDS.
This is no doubt due to the methyl group, since a similar
feature is noticed in the case of other methyl compounds.
It is, however, necessary to also take into consideration
other groups in a compound, and possible interactions.
Thus in the salts of formic acid we note the following
H . CO a CH,
. V m 627
17 x 3*7 = 62*9
A
- 0*2
H . CO a C 2 H 5
84*6
23 x 3 7 = 85*1
A
- 0-5
H . CO a C,H 7
106-2
29 x 3*7 = 107-3
A
- i-i
H . C 0 2 C 4 H 9
127*6
35 x 3*7 = 129*5
A
- i *9
H . CO a C 5 H n
150*5
41 x 3-7 = 1517
A
- 1*2
H . CO a C # H n
173*3
47 x 3*7 = 173*9
A
- 0*6
Thus the contraction increases as the group increases, up to a
certain point, and then diminishes. (See also curve.)
These effects are no doubt complex, and the exact analysis
of the data is difficult. Much depends on the exact configuration
of the molecules, and the particular groups acting. The first is
a subject about which comparatively little is known. Another
of the complications is, as already shown, the influence of sym¬
metry.
It will be noted that the series
(a) R—COOH aliphatic acids A
(b) R—COOCH, methyl esters
(c) R—COOCjH 5 ethyl esters
R—COOR' higher esters.
are truly homologous series, if we are to judge by the constancy
of the differences A between the corresponding members of pairs
of series.
On the other hand, if we arrange the substances as different
salts of the same acid, complications are introduced, if we are to
judge by the variable character of the differences A. For this
reason, these series are not regarded as truly homologous. By
the first method of arrangement, we notice that the group CH 8 ,
attached to the atom O, is responsible for a constant difference
A = — i*6, and is thus independent of the complexities of the
compounds. Similarly, the group C 2 H 6 is found to be responsible
for a smaller diminution. By comparing the volumes of isomeric
compounds we notice this depressing influence of CH a , QHg,
etc., both when attached to the O radical and the C radical, and
moreover the amount of the contraction can be estimated. This
method gets rid of the complicating influence of complexity.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 153
Any depressing influence of particular groups is also revealed by
an examination of the curves.
We obtain similar results from an examination of the di-
carboxylic esters, 47 but the depressions are comparatively small.
Compounds:
Methyl oxalate
V.
»ch 2 .
Or
(COOCH,),
COOCHg
“ 7*4
81 *o
2 X 18*2
dooc a H,
139*6
103*2
2 X 18*2
The groups COOCH s , COOC 2 H 6 have elsewhere been found
to be equal to 59 0 and 80*9 respectively. The numbers would
give to
(COOCH 8 ) 2 the. volume of 2 x 59*0 118*0 A= - o*6
COOCH 8
and | the volume of 59*0 + 80*9 139*9 A = - 0*3
COOC 2 H 5
(See Appendix 2, p. 264.)
The Unsaturated Oxygen Compounds. 48
/-
Acrolein CH, : CH • CT
\
b.p. 52*4 0*8410
C = 0*50 \m 69*9*
H
Propaldehyde CH S . CH 2 C
r
\
74*8
H
2H 7*4
CH 8 :CH.c' 67*4
A = 69*9 - 67*4 = + 2*5.
There is thus an augmentation due to unsaturation.
There are no data for acrylic acid, but Weger has studied the
esters of this acid.
Methyl acrylate CH 2 : CH . COOCH 8
b.p. 8o°. Vm 98*6
CH 8 . CH 2 . COOCH3 104 6
CH 2 : CH . COOCH, 97*2
A = 98*6 - 97*2 = + 1*4.
There is again an increase for unsaturation.
It is remarkable that a methyl acrylic acid
ch 2 : c;
ch 8
cooh
Digitized by v^ooQle
154
LIQUID CHEMICAL COMPOUNDS.
should possess a similar volume, viz. 98*4.
c.
Unsaturated esters:
Methyl acrylate, etc.
Vm.
Sat. Est.
-H a = 2 nV a .
A
0*515
CH a : CH . COOCH 8
98*6
104*6 (G)
97*2
+ i *4
o *595
CH, : CH . COOC 2 H 5
122*0
128*1 (S)
120*7
+ i *3
o* 55 i
CH a : CH . COOC 8 H 7
145*3
1510 (S)
143*6
+ 17
There is thus seen to be the same augmentation for all these
compounds. This is considered to be due to unsaturation, or
rather to the result of an interaction between an unsaturated
group and an ethenoid linkage.
The volume of the isomeric allyl acetate
CH 3 . CO . O . CH a . CH : CH a
is Vm 121*7.
Propyl acetate CH 3 . CO . OCH 2 . CH 2 . CH S
Vm 128*9
“ H a 7*4
2»V a 1215
Vm 121*7
There is no augmentation in this case, because the ethenoid
linkage is out of the range of the influence of the unsaturated
group -
O
O-
It has already been pointed out that the vinyl derivatives
(p. 88) CH 2 : CHX, etc., show the same augmentation.
Tiglic aldehyde
CH S . CH : C(CH 8 ). CHO
b.p. 116 d 16 0*871
Vm 109*5
2CH 8 52*0
2CH 37*0
CO 22*2
111*2
less for a & -3*1
108*1
for unsat. +1*5
%rNa 109*6
Vm 109*5
C = 0*52
5 c—C-CH 8 \
h/ II
CH—CH 8 /
Formula of tiglic aldehyde.
Ethyl angelate
CH 8 . CH : C(CH„). COOC a H 5
b.p. 141 do 0*9347
C 7 H, a O a CH 8 . CH a . CH a . CH a . COOC 2 H 5 (ethyl valerate)
C =* 0*578
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. 155
Ethyl angelate V** 163*9
2CHj 52
c 2 h b 48
coo 33*3
C 2 H 5 OOC-C-CH 3 *, t
C a H 33-3
166-6
CH -CH 8 /
less for a & struct. - 3*1
Formula of ethyl angelate.
163 s
for unsat. +1*3
The increase for unsaturation
—
is in this case doubtful. This
2»Va 164*8
Vm 163*9
may be due to inaccuracy in
the calculations.
Ethyl tiglate (ethyl and methyl crotonate)
CH, . CH : C(CH 8 ) COOC,H,
b.p. 152.
<*o •9425*
C ass *58 Vw 164*1
2CH 8 52*0
CjHj 48*0
COO 33*3
C»H 33*3
C 2 H 5 . OOC—C——CH 8 \^
CH—CH 8 /
166-6
less for a & struct. - 3*1
Formula of ethyl tiglate.
163-5
for unsat. +1*3
or
CH 8 —C—COOC 2 H 5
2»V fl 164*8
Vm 164*1
HC—CH 8
Ethyl angelate and tiglate are probably isomers of the type
H \ c . c / cooc » H »
H W CH *
CH,/ ' \CH,
CH,/ \COOC,H,
The evidence is not quite clear
as to which formula represents
the angelic acid derivative and which is the tiglate, owing to
want of precision in the data.
The Substituted Esters . 49
Ethyl monochloracetate
CH 2 C1 . COOC 2 H B V w
1239
CH 8 .COOC 2 H 5 Vm 1061
less H - 3*7
H
I
102*4
plus CF 21*5
Cl—c—c = 0
2 nVa 123-9
1 1
H OC 2 H b
Vf» 123*9
Digitized by CjOO^Ic
LIQUID CHEMICAL COMPOUNDS.
*5*
Ethyl dichloracetate
CHCl,COOC 9 H 5 Vm 143*9
CH,.COOC 9 H 5 V w io6*i
less H 9 - 7*4
98*7
plus Cl' +215
120*2
plus Cl" 23*5
2»V a 1437
Vm 143*9
Ethyl trichloracetate
CC1 8 . COOC 9 H 5
CH S . COOC 9 H 5 Vm 1061
less 3 H -11*1
950
plus cr _+21*5
116*5
plus Cl" 48*0 (2 x
2nVa 164*5
Vm 164*4
We find among the above compounds, that as the chlorine
content of the substances increases, the boiling-points are
diminished and the volumes augmented.
b.p.
A.
Va(Cl).
CH 9 C1 . COOC 9 H c
144*5
13*2
21*5
CHCla . COOC a H*
157*7
9*4
22*7 x 2
CC1 8 . COOC 9 H 5
167*1
23*1x3
This is exactly what was noticed in the chlor-substituted methanes.
It is not easy to say whether all the chlorine atoms partake
of the increase, or whether the increase is only operative on the
additional carbon atoms. An explanation of the fact of increase
is no doubt of fundamental importance.
We calculate the following values in order to obtain a broader
basis of fact:—
C.
b. P .
A.
^19*
Vm.
V(C1) average.
0*540
CH 9 C1. COOCH3
128*5
i5*5
1*2352
100*9
21*4
o*555
CHC1,. COOCHj
144*0
7*5
I*38o8
120*8
22*5 X 2
0*570
CCI3COOCH3
152*5
1*4892
141*1
23*0 x 3
Again, we notice the phenomenon of gradually diminishing
differences in boiling-points and increasing volumes as the
number of chlorine atoms increases. The values may be cal¬
culated from that of methyl acetate as before.
Cl"
I
Cl—c—c=o
I I
H OC„H.
Vm 163*4
Cl"
I
Cl'—C—COOC 2 H 6
Cl"
54*0)
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 157
Calculated
Esters.
2»V a -»H.
+ »C 1 .
values.
Vm.
CH S . COOCH 8
83*2
83*2
CH S C 1 . COOCH.
79*5
21*5
101.0
100*9
CHC 4 . COOCH 8
75*8
21*5+23*5
120*8
120*8
CClj . COOCH 8
72*1
21*5+48*0
141*6
141*1
A very interesting series of compounds is that of the mono-di
and tri-chlor acetic acids.
Monochloracetic acid CH a Cl. COOH
b.p. 186*0 d n 1*366 (Hoffmann).
C = 0*530 \ m 78*0
CH s COOH 640
less H - 3*7
plus Cl
2»Va
Vm
60*3
21*5
81 *8
78*0
We notice in this case a large difference of - 3’8 between the
two values. On examination we find that this is just what we
should expect, for on writing the formula
H
HX-
-Cl
■M
/
Cl
O: C
-O . H
y \
we see that the substance is an a /3 chlor-hydrin.
subject to a contraction of about - 3*1.
Trichloracetic acid. CCl a . COOH
b.p. 197*0 (mean) 1*617
OH
As such, it is
C = 0*570
64*0
Vm ii.9*5
3H -ii*i
Cl'
2CI"
less for aft structure
2«V«
Vm
52*9
21*5
74*4
48*0
122*4
-3-i
119*3
119*5
Cl
Cl-
C 1
“xl / 3
</\>H
Digitized by v^ooQle
158 LIQUID CHEMICAL COMPOUNDS .
Since there is only one hydroxyl group it can influence only one
chlorine atom, and so only the simple contraction — 3*1 is found
throughout the series.
The Dibrompropionic Esters . 48
Weger has determined the volumes of a number of dibrom
esters, which present some noteworthy features.
In order to facilitate the investigation we have calculated
the volumes of the corresponding a dibrom esters. For the
a /3 dibrom methyl compound C = 0 5 2 5
CH, . CBr, . COOCHg b.p. 182° d 0 1*9043 Vm 156*3
Vm I 56 * 3 *
CH a BrCHBr . COOCH, b.p. 205*8
Vm 152*4*
A -3*9
There is thus a considerable contraction for the a /3 compound as
compared with the other.
Starting from methyl propionate we can calculate the result,
remembering that the volume of an a bromine atom (from acid
group) is larger than one in the f 3 position with reference to this
group.
The f 3 bromine atom which is at the end of a chain only
possesses a volume of 27*5, whilst one in the af 3 has a larger
volume, 29*5.
CH, . CBr a . COOCH, Vm 156 3
CH 8 . CH a . COOCH, 104*6
less 2H - 7*4
97*2
2 Br" 59*0 (2 x 29*5)
2nV a 156*2 volume of a compound
Vm 156*3
CH a Br . CH . Br . COOCH, Vm 152*4
CH, . CH a . COOCH, 104*6
less H a - 7*4
97*2
CH9- CH - C = 0 Br' 27*5
I I I
Br Br OCH 3 Br
less for afi struct.
2 nV a 152*2
Vm 152*4
H Br
H—c—d— c = o
A L d)CH 3
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN . 159
The contraction - 3*9, already noticed, is made up as
follows:—
- 2*0 for ajS struct.
- 2*0 for A between a and & bromine
A -4*0
This may also be shown from an examination of dibrom
propyl alcohol.
ch 2
- Bn
\
1
1
1
/
CH-
I
- B K
V
I
)
ch 2 -
- OH'
3CH f 66-3
Br' 27*5
Br" 29*5
O' 7*4
1307
for ajS struct. - 6*2
2»V a 124*5
V m 124*3
ajS dibrom propionic methyl ester.
CH 2 Br . CHBr . COOCH s
CH s Br . CHBr . COOC a H ft
CH a Br . CHBr . COOC 8 H 7
The A for CH 2 is thus an unusually large number for com¬
pounds like the above. The only explanation is that there is an
augmentation for complexity. Thus the equivalent volume in
178 *6
CH 2 units is ^ y - = 8. We have noticed that compounds of
the order of complexity of octane (C 8 H 18 ) are subject to augmen¬
tations for the above reason. Elsewhere the augmentation per
CH 2 increase has been shown to be equal to +3*5 in certain
classes like the esters. Frequently the augmentation starts from
the methyl ester.
Vm. A.
152*3
26*3
178*6
26*0
204*6
Digitized by v^ooQle
160 LIQUID CHEMICAL COMPOUNDS.
a
dibrom propionic ethyl ester
CH, . CBr 2 COOC.,H 5
Vm 1817
a/8
CH a Br . CHBr . COOC 2 H B
178*6
A -3*1
CH S . CH 2 . COOC 2 H b
1277
less 2H
“ 7*4
Br
|
120-3
ch 3 — c-
-C = 0 2Br"
j
59-0 (2 x 29-5)
ir
oc 2 h b
i 79‘3
.
augm. for complex.
+ 3 5 (1 x 3 * 5 )
182*8
V m
1817
CH, . CH 2 . COOC 2 H b
1277
less H 2
- 7*4
riTT pTT
n A
OUo" — vU
--o
1 1
I
120*3
1 1
1 Br '
27*5
Br Br
OC2H5
147*8
Br"
29*5
177*3
less for a/8 struct.
— 2 *0
175-3
augm. for complex.
+ 3-5 (1 x 3-5)
2«V a
178-8
V w
I78*6
'
CH 2 Br . CHBr . COOC 8 H 7
204*6
CH, . CH 2 . COOC,H 7
149*9
8 less H 2
- 7*4
CHo-CH—
c=o
1 1
I
142*5
1 1
1 Br' + Br"
57 *o
Br Br
V._^
OC3H7
199*5
less for a/8 struct.
-2*0
197-5 ,
augm. for complex.
+ 7 0 (2 X 3-5)
2nV a
204*5
Vm
204*6
It is remarkable that neither the : O nor the OCH, groups
cause contractions.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 161
The Volume of :0 in Union with S, N, and P.
It has been shown that the volume of : O in union with carbon
varies from 7*4 to n*o.
Its volume, when the atoms to which it is attached are N, S,
P, is however different—uniformly so far as we can see
DO] - 8-2.
This is illustrated by a number of compounds which have been
investigated by Thorpe, but since their full consideration demands
a knowledge of the volumes of nitrogen, phosphorus, and sulphur,
this is left till later. We content ourselves at present with one
or two calculations which do not involve this knowledge.
POClj 101*4
PCI, 93-3
A for : O 8*i
107*4
992
8*2
S 0 2 C 1 2 863
S 0 C 1 2 780
A for : O 8*3
These results are sufficient to show that the volume of : O in
these compounds is 8*2. The other cases which involve a know¬
ledge of the volumes of nitrogen, phosphorus, and sulphur give
similar results.
POBrCLj
PBrCl a
A for : O
The Aromatic Acids. 61
When the radical C^H 6 replaces the aliphatic radical R, a
large contraction occurs.
Benzoic acid.
\
CH
CH
CH
/
C
Vm 126*9
C 6 H 5 92*8
COOH 34*i
- CbOH
A
- — 4 ’ 2 *
II
c b h u cooh
Vm 152*4
C 5 H u 114*1
COOH 38 3
Digitized by v^ooQle
162 LIQUID CHEMICAL COMPOUNDS.
Phenyl propionic acid and cinnamic acids.
CH
CH
^ \
✓ \
CH CH
CH CH
1 II
1 11
CH C—CH,—CH^—COOH
II
CH C—CH =
= CH—COOH
\ /
\ /
CH
CH
V w 170*9
Vm
162*7
C # H 5 . CH, . CH, 135*6
C 6 H 5 . CH : CH
127*5
COOH 35*3
COOH
35*2
A = - 3*0
A = - 3 *i-
The volumes of the residues C 6 H 6 . CH 2 : CH
2 . and CjH 6 .
' : CH . have been obtained as
follows :—
Ethyl benzene.
Cinnamene (styrolene).
C,H 5 . CH, . CH,
C 6 H 6 . CH :
CH,
Vm I 39'3
Vm
131*2
less H 3*7
less H
37
C 8 H 6 . CH, . CH, 135*6
C 6 H b . CH : CH
127-5
The fact that the two values
of COOH are
similar, even
though the saturated and unsaturated compounds are different in
constitution, is sufficiently remarkable.
Styrolene is made up as follows :—
CgH 8 92*8
CH : CH 2 407
2 nVa 133*5
less for unsat. and assoc, with C 8 H 5 - 2*3
C 6 H 5 . CH : CH a 1312
We thus work out the value for cinnamic acid
C 6 H 5 . CH : CH . COOH
C 6 H b 92*8
2CH 37*0
COOH 38*3
168*1
less for unsat. and assoc, with C 6 H S of hydrocarb.CH : CH - 2*3
165*8
less for assoc, of C 8 H 6 and COOH - 3*1
2#V a 1627
V m 1627
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 163
These numbers exactly agree. We are, however, met with
the difficulty of the possibility of interaction between two un¬
saturated groups QH 6 and COOH, when these groups are, on
the basis of the ordinary formula, separated from one another.
This difficulty can, however, be met by effecting a change in
this formula, whereby the two groups are brought near each
other. We may at the same time deal with phenyl propionic
acid. The suggested formulae for the two compounds are thus
CH 2 -CH 2 CH = CH
C 6 H 5 COOH C 6 H 5 COOH
Phenyl propionic acid. Cinnamic acid.
A = - 3*0 A = - 31
c 6 h 5
CH : CH
Aooh
Allocinnamic acid (possibly).
These results thus agree with those obtained by ordinary
chemical methods. There is a necessity for at least two modifica-
ti ons of cinnamic acid on the basis of the modified formulae.
Aromatic Esters.
It has been shown that in the aromatic acids the association
of QH 6 and COOH or other acid radical is responsible for a
contraction
a * -3 0.
In methyl benzoate C,H 5 . COOCH,
CH
HCOOCH,
Vm = 627
H 37
59 ‘o
* This difference may he in part explained by the fact that in methyl formate we
have the combination H - C - OCH„ in benzoic acid C - C - OCH s . If we compare
the latter with methyl acetate CH, - CO - OCH, which has a similar grouping , we
II *
CH
x
CH
CH C—COOCH,
\ /
CH
V w = 150*3
CgH 6 92*8
COOCH,
57*5
Digitized by v^ooQle
164
LIQUID CHEMICAL COMPOUNDS.
Ethyl benzoate C 6 H 8
. COOC a H 8
HCOOC a H 8
174*2
Vtn
84*6
c 6 h 8
92*8
H
37
COOC a H 6
81*4
80*9
A = -*0*5.
We thus see that the association of C*H 5 and the following
groups produce the contractions
A
rcooH -30 o— c=o
C 6 H 5 | COOCH 8 -i *5 II
lCOOC a H 8 -05 R C 6 H 5 .
The effects just noticed depend
{a) on the C*H 6 group;
( b ) on the nature of the R group;
(c) possibly on the circumstance that the neighbouring
carbons of the nucleus are unsaturated.
The group OH by itself occasions a contraction of only 0*8, so
that the large contraction for benzoic acid must be due to the
large amount of residual affinity connected with the group
Amyl benzoate C 6 H 5 .
c 8 h 5
co 2
92-8
333
114*1
240*2
augm. for complex + 7*0 (2x3 *5) for C 8 - C 4
5»V a 247*2
Vmt 247*3
The hydrocarbon chain of greater length than C 4 occasions an
expansion of +3*5 for every CH 2 added—
65 4321
—C—OCH a . CH a . CH a . CH a . CH 3
II
o
The volume of every member of this group of aromatic esters is
thus easily calculated.
find for COOCH, 83*2 - 25*9 = 57*3, which is similar to that for the same group in
benzoic acid. The volume of this group in ethyl benzoate C 6 H 5 . COOC 9 H 5 is about
1*5 higher than in methyl benzoate , as it should be
•COOCHg = . COOC 9 H 8 - CH a = 81*4 - 22 = 59*4. A = + 57*5 - 59*4 = - 1*9.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. 165
The Phenyl Propionic Esters.
CH
^ \
CH CH
I II
C C—CH, . CH, COOR
\ /
CH
In synthesizing the values of these compounds we utilize the
general data:—
C„H, 92-8 CH, 22-1 CH, 26-0 C,H, 48-0 C,H, 70-0
The augmentation for complexity above C 4 at A = + 3 ‘°-
Phenyl propionic propyl ester
C 6 H 8 . CH 2 . CHj . COOC 3 H 7
C 6 H 6 92*8
2CHj 44 ‘2
CO, 333
C s H 7 70-0
2403
augm. (C 6 - C 4 ) + 6-o (2 x 3*0)
2»V a 246*3
Vm 246-0
Vtn .
C 6 H 5 . CH 9 . CH 2 . COOCH3 195*2 196-3
C 6 H 8 . CH 2 . CH a . COOCjH 5 221-5 221*3
C 6 H 8 . CH a . CH 2 . COOC 8 H 7 246*0 246-3
A.
- i-i
The contraction - ri in the case of the first compound agrees
with former results.
The Cinnamic Esters.
CH
S \
CH CH
I "
CH C—CH : CH . COOR
\ /
CH
In these compounds the contraction seems to have disappeared,
for on utilizing the following data:—
C 6 H 8 92-8 CH 18-5 COOCH3 59-0 COOC 9 H 8 80-9 COOC 3 H 7 10,
numbers similar to those obtained by experiment are calculated:—
Digitized by v^ooQle
166 LIQUID CHEMICAL COMPOUNDS.
Cinnamic methyl esters
C 6 H b . CH : CH . COOCH3
C-H b . CH : CH 2
131*3
C«H 5
92*8
less H
3*7
2CH
370
127*6
COOCHg
59 *o
COOCH,
59 *o
SnV a
188*8
2»V a
186*6
Vm
188*3
V m
1883
A =
+ i* 7 h
There is thus an augmentation of 17 approximately equal to the
effect for unsaturation.
Compound.
V m .
2»V*.
A.
C 6 H 6 . CH : CH . COOCHj
188*3
188*8
-0*5
C 6 H 6 . CH : CH . COOC 2 H B
213*75
213*7
C 6 H b . CH : CH . COOC 3 H 7
239*4
238*8
—
Methyl salicylate C 6 H 4 (OH)(COOCH 8 ).
V m 1567.
CH X C-OH
CH ,C- -COOCH3
T 5 H
C B H 4 89*6
COOCHj 59*0 in open-chain compounds.
OH io*i in alcohols.
158*7
for assoc, of OH and C 6 H 6 - 0*8 (see for phenols)
157*9
for assoc, of COOCH 8 and C 6 H 6 -1*5 (see for methyl benzoate)
2nV a 156*4
V m 1567
* Such an increase is noted in methyl ethyl and propyl acrylates when compared
with the same salts of propionic acid :—
propionate
A
acrylate
OCHj 104*6
5-9
98*7
OC,H, 1277
57
122*0
OC,H, 1510
5*8
145*2
A
5*8 for H 2
A = 7*2
- 5*8 = + i ’4
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. 167
This is better seen from the following :—
V m
C 6 H 6 OH 102*0
C 6 H r . CO a CH s 150*3
Sum 252*3
less for C 6 H 6 - 96*0
2»V a 156*3
V m 1567
The volume of methyl salicylate can thus be accounted for
without supposing that there is any action between the two
groups in the ortho-position. That the association of these two
groups should normally interact is rendered probable by the
following calculation :—
0 Cresyl methyl ether.
-OCH3
y c \
CH C-
CH .C-
-ch 3 .
c 6 h 6 *ch 8
p cresyl methyl 148*0
o „ „ 146*4
A = - i*6
by summation—
A = - 1 '3
125*6
ii8*3
Sum 243*9
less C 6 H 6 —96*0
147*9
W m 146*4
A +i*5 for 0 structure.
This compound is evidently " adjacent ”
The only explanation which we can offer is that methyl
salicylate is an “ opposed compound.
OH
CH=CH I
/ \l
CH C
\ ✓
. CH-C
:ooch s
There is thus an “ adjacent ” modification also possible. A
number of exceptions to the rule among the different compounds
have been found, and since it is shown that their volumes do not
involve any contraction although ortho compounds, we conclude
that the groups are outside the range of each other's action. This
Digitized by v^ooQle
168
LIQUID CHEMICAL COMPOUNDS.
occurs under the circumstances that the groups are on opposite sides
of the nucleus, that is the compounds are “ opposed ” compounds.
The Volume of Ring Oxygen. 52
A preliminary study of the subject might lead us to suppose
that the volume of ring oxygen is invariable. A more minute
examination of the data shows that this is not so.
In the first place ring oxygen
is apparently ethereal in character, and as such might vary, even
as it does in
—ch.,— o— ch 3 - ch 2 —o—c 2 H 5
C 6 Hg—O—CH a .
Moreover, in certain cases, it is found that ring oxygen, or
indeed any other combination including oxygen, may be related
to the volume which it possessed in the original compound
If we consider such cases as
CH 2 -CH 2 -OjH
r J
GH<2- CH a -
CH.-CH,
\
/
O H
We see the possibility of ^)0 retaining its volume relative
to hydrogen in the ring compound.
A safe rule to follow is to assume the contraction compatible
with the class of ring we are considering, and then finding the
volume of oxygen by difference.
Ethylene oxide C a H O
V,* 50*0
CH aX
I >0 CH„CHO 567
CH a / 2CH a 44*4
> O = 12-4
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN. 169
2CH, 44-4
> O 12-4
56-8
less for O3 ring - 6 '°
5»V a 50*8
V m 50*0
CH S —O—CH,. CH . CH S V m 105*1
less C 4 H 10 96*0
>0 « 9*i
2CH 2 44*2
CH 185
' Cl 2i*5
> O 9*i
93*3
-6*o
87-3
87-3
Furfurane C 4 H g O Vm 73*2
CH = CH\ CH
ch=ch/°
, . CH a — 0 —
CH, . CH,
Vm
less C 4 Hjq
106*3
96*0
> O 10-3
4 ch
74 *o
>0
10*3
843
less for O5 ring
- iro
S«V«
73*3
Vm
73*2
Furfural C 6 H 4 O a
Vm 95*7
CH = C—CHO
4 CH
74 *o
N /°
>0
103
CO
22*1
/
CH = C
106*4
less for 0« ring
- n*o
2 «Va
954
Vm
957
less for Os ring
2nVa
Vm
Digitized by v^ooQle
LIQUID CHEMICAL COMPOUNDS.
170
Citraconic anhydride C B H 4 0 8
d u 1*2536 (Knops) b.p. 214
CH S —C-CCK C = 0*46
II JO Vm 106*4 by formula.
CH—CO/
This compound is a derivative of methyl succinic acid and
thus we might expect the volume of > O to be 7*4.
CH a 22*1
2CH 37*0
2CO" 52*0
>0 7-4
118*5
less for £7 b r » n S - iro
2nV a 107*5
Vm 106*4
Citraconic and mesaconic acids
CH 5 —C—COOH
j
CH—COOH
There is probably a contraction
for the attachment of the group
CH 8 to an unsaturated system.
cooh— c— ch 8
Ih—COOH
probably both possess similar volumes owing to the fact that
there are two groups in apposition in the two compounds. The
theoretical volumes are—
Citraconic anhydride 106*4
H 2 0 i 4 *8
121*2
less for a)9 struct. - 1*5
Va 1197
The whole contraction is -3*1, but seeing that there is an ex¬
pansion of +i*5 for the ethenoid link, the resultant contraction
is only — I * 5
Butyrolactone C 4 H 6 0 2
d Q 1*441 (Saytzeff) b.p. 206
C = 0*46 Vm 90*0 (by form.)
CHj—CH 2n ^
O
/
CH a — CO
3 ch 8
66*3
CO"
26*0
>0
9*6
101*9
less for £7 b rin g
-11*0
5 mV«
90*9
Vm
90*0
We have thus the following interesting series of compounds:—
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 171
CH,—CH,—OH ch 2 CH 2 -jOH CH = CH
CH a —CH 8
l
CHjCOOH
CH 2 -CH 2
Ih,-
-CH,
ch 2
\>
/
! CH=CH
O !H ^
1 Ring oxide (ethereal).
CH a —co x
CH-
CH 3 -CO |o H -* I
j CH,—CO^
or
CH-
Anhydride.
CO O | H
-CO
-CO
\
/
o
ch 2 —
■co ;oh]
CH,—CO v
I '0 Lactone.
1 /
CH 2 -
i !
1 —1 !
-
co o:hi
CH,—CH,
In every case the different peculiarities of structure can be stated
in terms of volume.
8 Hexylene oxide C 6 H 12 0
CH,—CH—CH.
/ \
CH, O
\ /
d 0 0-8739
b.p. 104*0 C = 0*46
CH 2 -CH, W m 129*2 by formula.
CH, . CH 2 .0 . CH, . CH, . CH,
c 3 h 7 —0
—C S H.
Wm 127*9
Wm
151*3
C|H m 1178
c«h 14
139*9
]> O 10*1
>0
ii *4
6CH,
132‘b
>0
n *4
144-0
less for £7e ring
-15-0
5 »V a
129*0
W m
129*2
Digitized by v^ooQle
172
LIQUID CHEMICAL COMPOUNDS.
Diethylene dioxide C 4 H 8 O a
CH a -CH a
/ \
\ /
CHj—CH,
d 0 1*0482
C = 0*46
4 ch,
2>0
b.p. 102*0
Vm 94*7
88*4
21*0 average 10*5
less for ring
SnV a
V m
109*4
-15*0
94*4
947
CH a -
Camphor C 10 H 16 O
-CH-CH.
CH
a-
CH,
CH.
Ah,
-C : O
V m 187*6
Ci 0 H i 6 2O 7 *2
: O n*o
218*2
less for ring - 30*6
2»V« *187*6
Vm 187*6
CH*
-CH-CH.
CH 3 —C—CH S
CH.
CH a
Vm 190*5
h 18
O' 6*4 (as in alcohols)
214*6
CH OH less f or r j n g _ jQ.g
S»Va 190*4
V m 190*5
Paraldehyde C 6 H ia O a
CH,
L
/ \
CH S —CH CH—CH.
\/
O
v« 151*0 (Schiff).
C«H„ 133*2
3 > O 33*0
166*2
less for /~~7 fl ring - 15*0
2 n V a 151*2
V m 151*0
Paraldehyde has a similar structure to that of mesitylene.
Digitized by
Google
ORGANIC COMPOUNDS CONTAINING OXYGEN 173
OH.
CH a
CH
/ \
o o
CHg—CH CH—CHg
Paraldehyde C 6 H ia 0 3
V« 151*0
✓ \
CH CH
CHg—C C—CHg
CH
Mesitylene C 6 H 12 C 3
162*4
A = -n*4
The difference in volume should be on the assumption of
similarity of structure.
A = 3C - 3O = 3(C - O) = 3(14-8 - n o)
= 3x3*8 = 11*4
This difference is identical with the one observed.
Summary of Chapter on Oxygen.
Oxygen has been shown to possess quite a number of values,
according to its position and function. This is in opposition to
Kopp’s idea that only two values are found :—
O' = 7*8 O" = 12*0
The reason doubtless is, that the volume of O and every other
atom varies somewhat according to its environment. Be¬
sides this, we ought to consider the nature of the forces acting,
that is the residual affinity. An exact notion of these circum¬
stances eludes us. All we can do is to tabulate the values, and
compare them with the known constitutions of the compounds.
Under favourable circumstances we can draw up a series of rules
which may act as guides in enabling us to decide on the real
volume of oxygen in an uninvestigated compound. The inquiry
has not gone quite far enough to do this with precision. In
molecular volumes we roughly classify the varying types of
oxygen as:—
(a) Hydroxyl oxygen
O'
5-6 - 7-4
(b) Ethereal oxygen
>0
9*0 - n*o
(c) Aldehydic and ketonic 0 "
7*4 - n*o
(d) Carboxylic
0.
16*4 - 18*0
Digitized by v^ooQle
174
LIQUID CHEMICAL COMPOUNDS
TABLE LXX. —Values of Oxygen in Organic Compounds.
O'
Aliphatic.
6*4 (in alcohols)
^>0 (ethereal)
O"
7*4 (in esters) or [n*o]*
9*5 (* n —OCH s )
io-ii (in —OR)
Aromatic, etc.
5*6 (in phenols)
7*4 (in esters) or [iro]*
7*5 (in —OCH s )
io-ii (in —OR)
aldehydic and
ketonic
O"
carboxylic
°"
in union with
S, P, N
7*4 (aldehydes)
n*o (ketones)
ii*o (in acids and esters) or [7*4]*
8*3
Ring oxygen as in ethers .
A requirement of the first importance is to know how the
volume of oxygen varies according to its position in a chain. In
some cases the information is definite and complete, in others it
is very indecisive.
H—O—R CH 8 —O—R C 2 H 6 —O—R C 4 H 9 —O—R
Va 6*4 9*6 10 11
These facts show that the volume of the terminal oxygen is
always the minimum value, and the volume increases as the
oxygen passes into the interior of the molecule.
Similar results are shown in the following:—
CH S —O—C —R C 2 H 6 —O—C —R C3H7—O—C— R
II II II
OOO
30*6
32*0
327
When the C 0 2 is in the middle of the chain the volume is a
maximum. Another way of stating the same thing is to suppose
that the groups CH 3 , C 2 H 6 , C 8 H 7 produce depressing actions on
the volume in an order which is in the inverse of the order in
which they are written down.
CH S . CH a . COOCHg Vtn 104*6 %Na 107*0 A -2*4
CH 3 . CH 2 . COOC 2 H b 1277 129*1 -1*4
CH S . CH 2 . COOC 3 H 7 151*0 151*2 —
That the methyl group is specially concerned in producing a
contraction is shown by the following—
CH3.COOCH3 Vm 83*2 SnVa 85*3 -2*1
Similarly in aromatic compounds there are contractions when
* See Note on page 176.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 175
two CH 3 groups are in the ortho position relative to each other.
By analogy we have the configurations :—
av - 2*4 - 2*4
A b. p. + 2 - 4 + 3 0 .
An examination will show that the boiling-points of the
methyl esters are always 2-4 degrees higher than what might be
expected from the differences.
For some reason the acids show an augmentation instead of
a depression, which fact is contrary to expectation.
Acid.
v m .
A.
Ester.
V w .
CH 8 . COOH
63-9
+ 1*2
hcooch 8
62*7
C a H B . COOH
88*4
+ 3*8
H . COOC 2 H 6
84*6
C 3 H 7 . COOH
107*9
+ i *7
H . COOC 3 H 7
106*2
C 4 H 9 . COOH
130*0
+ 2*4
hcooc 4 h 9
127*6
C 6 H 9 . COOH
152*9
+ 2*4
HCOOC 5 H n
150*5
The reason for this is :
not known.
We may also consider the question of the change in volume
under the following circumstances :—
CH a (OH) . CH a . CH a . CH a . CH 8 CH 8 . CH(OH). CH 2 . CH 2 . CH 8
CH 3 . CH, . CH(OH) . CH a . CH 8
CHO . CH a . CH a . CH a . CH 3 CH 8 . CO . CH a . CH a , CH 5
CH 3 . CH a . CO . CH a . CH S
The information on these matters is meagre and inconclusive.
A general rule seems to be that the volume of a typical atom
increases as it is placed further from the end of the chain, and
attains its maximum when it is attached to the central carbon
atom. Accompanying this is a corresponding depression in the
boiling-point of the compound.
As a result of careful examination we think it desirable to
exclude such general and vague ideas as symmetry, shape, etc.,
as possible influences operative in deciding the volume of a com¬
pound, and to substitute more precise notions, as the specific
actions or influences, due to particular groups. The forces acting
are doubtless those of residual affinity, and though symmetry,
etc., are involved, yet it is only indirectly.
The most probable explanation of the facts, and one which
is intiipately concerned with the question of what we mean by
molecular volumes, is as follows: As a result of residual affinity
Digitized by v^ooQle
176
LIQUID CHEMICAL COMPOUNDS.
arising from the molecule, there is produced an external field of
strength which may be the sum total of the effects arising from
individual atoms. This external field causes an intermolecular
pressure which at once influences the boiling-point and the
molecular volumes. If the field is relatively great, the boiling-
point is raised and the volume diminished by compression and
vice vers&.
Thus (a) the alcohols are more reactive than the ethers. We
find relatively high boiling-points and small volumes. The effect
of an oxygen atom diminishes as it tends towards the middle of
the chain.
(b) Similarly when we compare aa with a/8 compounds we
are comparing the effect of the combinations—CH 2 —CHX 2 and
—CHX—CH 2 X. As X becomes removed from the end of the
chain the boiling-point diminishes and the volume increases.
The two atoms X, however, may also influence one another, and
by their proximity when attached to different carbon atoms
intensify each other's action, and thus the external field. This
would account for the observed increase in boiling-point and
diminished volume as compared with aa compounds.
(c) The above considerations also affect ortho compounds as
compared with para, and from what we have seen the methyl
esters as compared with ethyl and propyl.
We should expect to find the capillary constants and viscosity
coefficients of the high boiling-point compounds relatively high.
This is what we do find.
Note.
Alternative Interpretation of tv a . (0 2 ) = 18 5.
It has been shown that 0 2 in the esters possesses a volume
0 2 = 18*5,
and O" = ii*o, O' = 7*4 (seep. 147),
which numbers are similar to those indicated by Kopp.
Whilst this book was passing through the press an alternative
explanation of the value of 0 2 suggested itself and it has much to
recommend it. The chief advantage is that thereby a considerably
greater consistency in theory is secured than by the former ex -
planation.
Digitized by v^ooQle
ORGANIC COMPOUNDS CONTAINING OXYGEN 177
Thus :—
O" = 7*4 (as in the aldehydes and ketones and doubly bound oxygen
generally),
O ' ^>0
O' = 9*1 (in methyl esters); g*g (methyl ethers ).
= n*o (in higher esters) ; 11*0 (higher ethers),
= 12*0 (in acids).
The value of O' is similar to that of ethereal oxygen , which from
a chemical point of view it is. The only difficulty is to account for
the value O" = n in COCl % and analogous compounds.
If the above be the true explanation , which we believe it is,
Kopp's theory, 0 ,f — 12, O' = 7*8, is found to be once again at
fault.
12
Digitized by v^ooQle
CHAPTER V.
THE MOLECULAR VOLUMES OF SULPHUR COMPOUNDS.
The Element Sulphur.
The molecular volume of sulphur in the uncombined state is
S = 21*8 Ramsay 52
and in the combined state
S| = 21*6
S, = 25*6
The first value in the combined state is for doubly bound
sulphur : S.
The second value is considered to be the normal one.
In considering the characteristics of sulphur at, and near the
boiling-point, we find—
fusion-point 111-5 yellow mobile liquid
„ 250*0 dark and thick
„ 300-0 dark but thin
boiling-point 440*0 orange-coloured vapour.
In the liquid state,
the molecules are probably S 10 and S 8 .
At any rate in the vapour, the complexity is
at 500° s # .
and at iooo° S,
There is thus dissociation
S* ? 3S a .
The formula for S 2 is certainly
<S = S>
but the molecule S 8 , so far as we know, has not been explained.
The formula for S„ must explain at least three facts—
(a) The existence of colour.
lb) The easy decomposability of S 8 into 3^.
(c) An average value of sulphur of 21*8.
178
Digitized by v^ooQle
MOLECULAR VOLUMES OF SULPHUR COMPOUNDS. 179
The colour phenomenon can be explained on the assump¬
tion that the formula for S 6 contains at least three double
bonds.
The average value for sulphur of 21*8 can be explained
either by an open-chain formula or by one which, as a whole or
in part, involves ring structure.
The open chain formula would be
: :
<s = s = s = s = s = s>
: :
j i
This explains colour and easy decomposability.
The possibility of this type of formula depends on whether
sulphur has sufficient power of self combination to form a
molecule with this length of chain.
That this may be so is suggested by a formula for potassium
pentasulphide
k-s-s-s-s-s-k.
We must also assume that = S = possesses the volume
of 21 # 8, if we take and consider the above formula.
The volume of sulphur in
o =« s = o
25*6 is contrary to this idea.
Possible ring formulae are—
Average volumes.
22*5 22*7 21*6
The only one which gives an average volume similar to the
experimental one is the third, but should be excluded owing to
the fact that easy decomposition is not suggested.
In spite of the fact that the value of S for the first, is slightly
higher than that indicated by experiment, it seems the most
satisfactory one. It explains the phenomena of colour and easy
decomposability.
12 *
Digitized by Google
i8o
LIQUID CHEMICAL COMPOUNDS .
I
S
The average volume of sulphur is calculated thus—
3 atoms of: S 3 x 25*6
768
3 „ S 3 x 21*6
64*8
6 atoms of S
141*6
less for ring
-6*5
S«
I35*i
s
22*5
Probably we do not know all the facts connected with a
formula of the above type. It is quite possible that there are
influences which would tend to lower the average value to the
one given by experiment.
The Halogen Derivatives of Sulphur (Thorpe, loc. cit.).
Sulphur di-chloride.
Cl—S—Cl.
b.p. 64
Vm
S
2C1
3«Va
Vm
do 1*62 C 0*46
68*8 (by formula)
25*6
43*2
68*8
68*8
Sulphur monochloride.
Sod,
Vf n go*3 (Thorpe).
Two formulae are possible—
Cl
/
s = s
\
and
Cl
Cl
/
1
Cl'
The first is the analogue of o « s , and the second is an
a# compound.
Digitized by v^ooQle
MOLECULAR VOLUMES OF SULPHUR COMPOUNDS . 181
ist formula. 2nd formula.
/
2S
512 (2 X
25-6)
= s
25*6
2CI
43' 2 (2 x
21*6)
\
= s
2C1
21*6
43*2
less for ajS str.
94*4
-3*1
2nV a
90*4
'ZtNa
9i # 3
v w
90*3
Vw
90*3
A - i*o
The validity of the first is suggested by the calculations.
One other compound like the above is
Sulphur monobromide.
S 2 Br 2 .
Br
/
S = S b.p. 170° d 4 2*628 C 0*46
\
Br
Vw ioo*o
/
= S
25*6
\
= s
21*6
2Br
54*0 (2 x 27*0)
5nVa
101*2
Vm
100*0
The calculations are found to be distinctly in favour of the
first formula, to which we adhere.
We possess also the following evidence for the atomic values
which have been proposed:—
SC1 2 68*8
SSCU 90*3
t S 21*5
It follows that
and
possesses a volume of 25*6
one of 21 *6
This is proved by the following results:—
Cl
S'V 256
C 2 H 5 —N = Ct=S
Vw 99*3
C,H 6 .N:C> 77-5
: S
21*8
Digitized by v^ooQle
182
LIQUID CHEMICAL COMPOUNDS.
Thorpe has come to the same conclusion, but in a different
way. Thus:—
SSCL, Vm 9°*3
ci.
45*4 (2 x 227)
2S
45*0 (2 X 22*5)
2nV a
9°*4
V w
90*3
We have seen that the two sulphur atoms are not equal in
value, but possess respectively the two volumes characteristic of
this element Moreover, we have reason to think that the
chlorine atoms do not possess quite so large a volume as Thorpe
and, indeed, Kopp supposed.
Thorpe has investigated a number of compounds which are
similar to the above, but containing oxygen, and the present
theory affords valuable evidence as to their structure. The con¬
clusions arrived at more than thirty years ago(i88i) are con¬
sidered to be invalid They were obtained by means of Kopp’s
numbers. The latter thought that only two values for oxygen
are met with,
O' 7*8 and O" 12-0.
This is not the case. At any rate, the particular solution of the
problem arrived at here hinges on the volume of oxygen.
SCI,
68*8
SOCL,
78*0
S0C1,
78*0
SOjCl,
86-3
O"
92
O"
8-3
Similar volumes are found when oxygen is united to phos¬
phorus and nitrogen.
PCI3
93*3
NO«
32*0 N
15*6
POCI3
101*4
20
16*6
0"
8*i
-ir
3«Va
32*2
T ■ IT
^0
Vm
32*0
The volume of oxygen in thionyl chloride is apparently
slightly larger than the other value. This is also the case for
one of the oxygens in sulphuryl chloride and other compounds.
The reason for this is not very apparent, but it at least explains
the volume of sulphur dioxide S 0 2 .
Digitized by v^ooQle
MOLECULAR VOLUMES OF SULPHUR COMPOUNDS . 183
so a
O
II
CO
II
0
Vm
43*9
= S =
25*6
2O"
18*4 (2 x 9*2)
3*V«
44*o
Vm
43*9
If this be so, then the volume for chlorine is 21*6.
SCI* 68*8 2CI = 68*8 - 25*6 = 43*2.
Cl 21*6.
This conclusion may be arrived at in another way—
S0 2 Cl(OH) 75*05
SO a Cl = 75 - 10*4 = 64*6 (OH * 6*7 4- 37)
SO a Cl 3 863
SO s Cl 64*6
Cl 217
Thus—
S = 25*6 O" = 9*1 and 8*3
Utilizing these values, we find—
Cl =■ 21*6
Thionyl chloride.
so*C4
Cl—S^-Cl
4
'XtNa 78*0
Vm 78*0
Sulphuryl chloride. Sulphuryl chlorhydrate.
SO a Cl(OH)
O
II
Cl—S—Cl
II
o
86*3
86*3
O
: JL,
A
75* 1
750
OH
The latter compound by condensation gives disulphuryl
chloride
Cl-S-OlH+HoV-S-Cl = Cl-S-O-S
O
' 1
- ClSO a » C1S0 9 0H—OH
= 75*0 - io*4^= 64*6.
The volume of —O— is thus—
Volume of ClSO a — O—O a SCl
2C1SO*—
133*5
129*2
-Cl
4*3
Digitized by v^ooQle
184
LIQUID CHEMICAL COMPOUNDS.
This is a small value for oxygen, indeed, the smallest value
known.
A possible explanation of this lies in the fact that in the
aromatic hydroxy compounds like phenol C 6 H 6 OH, where OH
is united to the nucleus C 8 H 5 _, there is a contraction of - 1*4.
E.g. if C 6 H 5 OH possess the volume 102*0
and C 6 H 6 „ „ 96*0
then O' 6*o
Since O' = 7*4 A is 6*o - 7*4 = - 1*4.
Now the —O— atom in disulphuryl chloride is joined on
to two acid groups —S 0 2 C 1 , or, at any rate, to groups pos¬
sessing residual affinity. If we deduct 2 x 1 *4 from the normal
value for this kind of oxygen we obtain the volume
O = 7*4 - 2*8 = 4*6.
This number is not very different from that found, 4 3, and thus
its remarkably small volume may possibly be explained.
The formulae supported by the molecular volume theory, and
which are considered to possess doubly bound oxygen, are dif¬
ferent to those which Thorpe has proposed, based on the con¬
clusion that the oxygen atom is of the hydroxyl type. Such
formulae would be
ci-s—o—ci, ci—o—s—o—ci
Thionyl chloride. Sulphuryl chloride.
In addition to the fact that the atomic volume of oxygen found,
in spite of its similarity to the value for one similar to —O—, is
a doubly bound one, a formula of this type could hardly apply
to sulphuryl chlorhydrate, for
Ci—o—s—O—OH
is extremely improbable.
The formulae which have been proposed are also more in
accord with the properties of the compounds and their genetic
relationships. It should also be observed that they support the
view that sulphur is tetra and hexavalent in the compounds
rather than divalent.
The Sulphur Oxides, Acids, and Ethereal Salts of the Acids.
These oxides are two in number, S 0 2 and S 0 8 .
Sulphur dioxide.
O = S = O.
Vm 43 * 9 *
Digitized by v^ooQle
MOLECULAR VOLUMES OF SULPHUR COMPOUNDS. 185
Oxygen appears to have the volume 9*2 in this compound
s
25*6
20
18*4
SnVa
44*o
Vm
43*9
This is also seen in thionyl chloride,
o
Cl—S—Cl
Vm
78*0
V
VmSCI 2
68-8
: 0
g-2
Sulphur trioxide.
S0 3 .
This compound possesses a molecular volume but little
different from that for S 0 2 . There must consequently be a
considerable contraction.
Vm 44*3
S 25*6
3O 249
5°*5
Vm 44*3
A -6*2
This contraction is evidently connected with the presence of a
three-membered single ring.
The formula to be accorded to this compound is for this
reason considered to be represented by
/
o
o = s
\
o
so that sulphur is tetravalent instead of hexavalent. It is re¬
markable that none of the groups S 0 2 , N 0 2 , CO a contain this
ring. The formulae are
O = N = O, O = S = O, 0 = 0 = 0,
and not
/
—N
\
** V.
/
/
» —s
. <C
\
\
Digitized by v^ooQle
i86
LIQUID CHEMICAL COMPOUNDS.
b.p. 161 0
Sulphurous ethyl ether.
(CjH 5 0 )S 0 .
^ i*io6 C 0*50 as in (C,H b O)CO
Vw 146*5 (by formula).
\
s
/
; O
c 4 h 10
96*0
s
25*6
30
24*9
3 »Va
146-5
Vm
146-5
o
Methyl sulphate.
(CH s O) 2 SO a
b.p.
188 0 d 16 1-333 C
V w ii2*2 by formula.
2CH 3 52*0
SO a 42-2
o*
18-6
0*50.
2»V tt III*8
V m 112*2
Ethyl sulphate.
(C*H 5 0 ) a SO a
b.p. 208° d ia 1*1837 Vm 155*7
2nV a 156*8
Doubly Bound Sulphur.
Sulphur, as already indicated, has two values. The second
has just been found to be associated with the ring compounds
thiophene and methyl pent thiophene. It is also found in the
thiocarbimides and thiocyanates.
Compound.
Vm.
S.
Vm.
The Cyanides.
CH». S . cn
78*2
21*9
56*3
CH 3 . CN
C a H 5 . S . CN
IOO’I
21*8
78*3
C a H 5 . c . n
c 6 h 6 . s . cn
143*4
21*8
121*6
C 6 H 5 . CN
A similar number is found to be connected with doubly
bound sulphur.
If sulphur mono-chloride possess the formula
/ C1 / C1
S = S , analogous to O = S
V, \
XI
Cl
the volume of : S would be 21*8 (q.v.) whilst that of = would
be 25 *6.
Digitized by v^ooQle
MOLECULAR VOLUMES OF SULPHUR COMPOUNDS. 187
Phosphorus sulpho chloride.
/ C1
S - P—Cl .
. Vm Ii6*i
/ Cl
<P—Clj»3-3
^C1
S 22*8
^Cl
Carbon bisulphide. .
S = C = S
V» 62*1
C 14*8
2 ; S 47*2
SrtV a 62*0
Vm 62*1
The reason is probably that
the volumes of sulphur are
slightly larger than usual, viz.
23 6 by analogy with CO a . This
idea is supported by the value
found for sulphur in PSC 1 3 22*8.
Ring Compounds including Sulphur.
The compounds answering to this description are few in
number. They are
CH==CH
S
Thiophene C 4 H 4 S M
V m 85-1
4 ch
74*0
s
22*0
96*0
less for ring
-11*0
2nV a
85*0
Vm
85*1
/
CH a
\
CH,-C:
fi methyl pent thiophene C 6 H 8 S
V
S
/
CH
113°*°
°*9935
0*48 like toluene
Vm 125*9
C,H 8 118-4 (32 x 37 )
S 22*0
I 4 0*4
less for £7« -15*0
2nV« 125*4
Vm 125*9
It is noteworthy that sulphur possesses a volume of only 22*0
in these compounds and not 25*6 as we should expect from a
study of the thio ethers. This is different from the results
obtained for oxygen and nitrogen.
Digitized by v^ooQle
LIQUID CHEMICAL COMPOUNDS,
188
The Thio Alcohols (Mercaptans) and Thio Ethers.
R—S-—H R—S—R
The available data are not very numerous, but the conclusions
are fairly certain.
The mercaptans.
Compound.
v m .
R.
S.
C a H # . S . H
77*6
52*0
25*6
C 5 H u . S . h
143*6
117*8
25*8
ch 8 . S . CH-
c 2 h 5 . S . C 3 H 6
The thio ethers.
121*6
96*0
25*6
It is remarkable that the value for sulphur just indicated is
also found in very dissimilar compounds, which include di-, tetra-,
and hexavalent sulphur in their composition.
It may further be remarked that the sulphur atom does not
seem to exert any special influence on the alkyl radicals, that is,
in modifying their atomic volumes, as compared with those found
in the normal paraffins. Moreover, no distinction is discoverable
between the mercaptans and the thio ethers in regard to their
molecular volumes. This is quite different from the results
obtained in the series
R— 0 —H
R—O—R
R—NH 2
R a =N-
-R
S,0C1 4
b.p. 100
d 1*656
C = 0-45
cn
rci
Vm 147 '°
Cl ts—s-
1
2 S =
51-6
ClJ 1
1=0
4 C1
88-4
0
8-3
SnVa
148-3
Vm
147-0
A
-1*3 for afi structure
tetrachloracetyl oxide
V
2C
4 C 1
o
* a
Vm
A
-3*0 for ajS structure
This is analogous to
Digitized by v^ooQle
MOLECULAR VOLUMES OF SULPHUR COMPOUNDS. 189
It has been found that there is an increase in volume of + 1 *5
in —CC 1 : CC 1 —and similar groups, so that the apparent value
of A is halved. We may suppose that S 2 . may be represented by
/ \
and that sulphur is hexavalent. We do not, however, think that
this is the case.
o
b.p. 179*5
Cl
/
\
Cl
Selenium Se
Selenyl chloride.
SeOCl*
<*i3 2 ’443
: Se^
.'i
CL
C 0*45
V m 79*0
27*0
9*1 see thionyl chloride
43* 2
2nV a 79*3
V» 79*0
Selenium chloride
b.p. 169°
Cl
/
Se — Se
\
Cl
Se^Clg
d 17 2*906
2CI
- Se
C 0-45
V« 91-4
27*0
43*2 (2 x 21*6)
21*6
Br
/
Se = Se
\
Br
2 nVa
91*8
Vm
91*4
Selenium bromide
SegBr a
V m 104*8
.. /
= Se
27*0
\
= Se
21*6
2Br
56*0 (2 x 28*0)
104*6
Vm
104*8
The above three selenium compounds are quite analogous to
the corresponding sulphur compounds. For the reasons stated
the formula is
Digitized by v^ooQle
190
LIQUID CHEMICAL COMPOUNDS.
excluded. Moreover, selenium is perfectly analogous to sulphur
in possessing two values, as must be the case from the nature of
the formulae.
Thus, the first value for Se x is 27*0 S x 25*6
and the second value for Scj 21*6 S a 21*6
The second is similar in value to that of sulphur.
Chromium Cr.
(Thorpe loc. cit.)
Not many chromium compounds have been investigated.
Indeed, the only one is chromyl chloride Cr 0 2 Cl 2 .
Cl
Chromyl chloride.
CrOXL
Vtn
88*2
2CI
43*2
2O
ie-e
Cr
28*4
2»Va
88*2
V m
88*2
Cl
Cl
Chromyl chloride seems perfectly analogous to sulphuryl
chloride, and chromium differs from sulphur in possessing a
slightly larger volume.
S 25*6 Se 27*0 Cr 28*4
The volume of tellurium is not known, but probably it pos¬
sesses one of from 37-0 to 40*0.
Elements of
Group 5.
M.W.
Vm .
Oxygen O
16
7*4 under and over
II-I2
Sulphur S
32
21*6
25*6
Chromium Cr
52*5
28*4
Selenium Se
78-9
27*0
Digitized by v^ooQle
CHAPTER VI.
THE MOLECULAR VOLUMES OF NITROGEN COMPOUNDS.
THE atom nitrogen, perhaps, not even excepting oxygen, is
of greatest interest to us, in consequence of its presence in many
interesting and useful compounds.
The element nitrogen
N = N
Nitrogen being a low boiling-point substance, it may be
useful to calculate its value by D. Berthelot’s method.
T c
127 0 Pc 33 atmos.
Vm = ii‘i x I27 a
D 33 x (254 - 78*6)
Tb.p. 78*6
= 30 * 9 .
The observed value is
Vtn 30*9 = 2 x 15*5.
Nitrogen, like phosphorus, is typically a trivalent element,
and if we are to judge from the latter, the direction of its
valencies when unconstrained are, as in the diagram, along
the edges of a tetrahedron.
Cl
Such a disposition accounts for the possibility of both nitro¬
gen and phosphorus functioning as pentavalent elements. On
comparing the atoms carbon and nitrogen in this respect, we see
that carbon cannot assume any higher valency than four—it is
saturated, whilst nitrogen may take on two more valencies—it is
unsaturated.
191
Digitized by
Google
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS . 193
It is remarkable that the above value for chlorine is found in
such compounds as R—Cl and CH 2 C 1 2 . In this, nitrogen
chloride resembles PBr 3 , that is, it possesses the minimum value
for the halogen element found in ordinary compounds. The
volume of bromine in PBr 3 is also that found in the alkyl brom¬
ides R—Br.
The Cyanides or Nitriles.
R—C = N.
These compounds belong to the class known as unsaturated,
and it will be found that the volume of nitrogen in them is
normal.
TABLE LXXI.— The Volumes of the Alkyl Cyanides.
Compound.
Vm.
^Va(R).
CN.
b.p.
(CN) 2 ....
60*4
—
30*2
21*0
CH3.CN .
57*3
26*0
3 i # 3
8i*6
C 2 H 6 . CN .
78-5
48*0
30*5
98*0
C 8 H 7 . CN .
100*0
70*0
30*0
118*5
Mean value .
30*5
The group CN resembles the combination N 2 in its constitu¬
tive features, and might thus be expected to show a similar
volume in the case of combined nitrogen.
—C = N N = N
C i N = 14*8 + 15*5 = 30*3.
It has already been shown that
N ; N = 30*9 = 2 x 15*5.
Isomeric with the nitriles, we find the carbylamines or iso¬
cyanides.
R—N = C >
In these compounds, carbon appears to be diatomic as in
<C C = O. They will be considered after the amines.
Cyanic Acid and Esters.
RCNO.
The compound is only stable below o°.
13
Digitized by v^ooQle
194
LIQUID CHEMICAL COMPOUNDS,
The boiling-points of the esters are
ethyl isocyanic ester b.p. 6o°
methyl „ „ „ 44 0
acid ,, 11 11
It is concluded that the boiling-point of cyanic acid is be¬
tween o° and 2 5 0 .
d - 20
1*1558
do
1-1400
dao
1*1242 by extrapolation
Vm 38*2 at 20 0
Vm 377 at o°
H
37
CN
30-2
O
8*3
2 «V«
42*2
Vm
377
A
-4-5 or -4-0
A contraction of this nature suggests that there is a three-
membered ring in the molecule of cyanic acid. If we con¬
sider the possibilities of the empirical formula HCNO, we find,
on starting from hydrocyanic acid and the hypothetical acid of
the carbylamines, that the following scheme may be drawn up.
In the case of both non-oxygen compounds, the possibilities
involved include a ring compound. The contraction associated
Digitized by v^ooQle
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS . 195
with so-called cyanic acid would lead us to suppose that it has a
different structure from that indicated by the ordinary formula
(marked 11 ). The only possible one, according to the data, is
H—C = N >
\/
o
The instability of the compound, and the tendency to modi¬
fication, might lead us to suppose that the ring compound is
transformable into the open chain one and vice versd. When
however, the hydrogen is replaced by an alkyl group, the latter
fixes the constitution and the cyanetholins result. The transition
to the second group of compounds, chief of which are the iso¬
cyanates, probably demands and points to an “ intermediate ”
compound, involving ring structure. This condition is satisfied
by Formula I for cyanic acid.
Thio Compounds.
There are, corresponding to the oxygen compounds, the
following combinations with sulphur.
N = C—SH and < S = C = NH
Thiocyanic acid Isothiocyanic acid
Sulphocyanic acid Sulphocarbimide.
Hydrocyanic acid HCN at the boiling-point 27 0 possesses a
volume of 38-8.
HCN 38*8
CN 30-2
H 8-6
This enormous volume for hydrogen is what we should
expect. On this basis the volume of CNSH would be
HCN
S
38*8
22*0
HCNS
60*8
Clasen gives the density of (^ 0 ) 1-0013 for what is called
hydrosulphocyanic acid
or Vm 58 9
so that we see that the constitution is that of an open-chain
compound, perhaps HS . C = N. However this may be, thio-
13 *
Digitized by
Google
196
LIQUID CHEMICAL COMPOUNDS.
cyanic acid gives rise to esters, which have been well studied,
and which possess the formula
N = c— SR
The hypothetical HN = C = S > has derivatives called
thiocarbimides R—N =* C = S >
The sulphocyanides.
Methyl sulphocyanide.
Ethyl sulphocyanide.
cHjS. c :
N Vm 78*2
c 2 h 5 s . c
l N
ch 8
26*0
c 2 h 5
48*0
s
22*0
s
22*0
CN
30-2
CN
30*2
2nV a
78*2
SnV a
100*2
Vm
78*2
Vm
100*1
V m 1001
b.p. 146
We see that these compounds give results in perfect accord
with their formulas.
Isothiocarbimide CS : NH is not known, but some of its
esters, the mustard oils, S = C = NR are.
We only give details of the ethyl ester in this place; the
others are dealt with after the amines for reasons to be given
later.
Ethyl thiocarbimide.
C 2 H 5 N : C : S b.p. 133° V m 99 3
Vw 99*3
2»Va IOO-2
-0-9
There is thus a difference of 13 0 between the boiling-points
of the two isomers, and a difference in volume of 0*9. The
thiocarbimides possess the lowest boiling-points and the smallest
volumes.
The carbylamines give rise to the isocyanates or carbimides.
Ethyl carbimide C 2 H 6 —N = C = O
b.p. 6o° di 0 0*8981 C = 0*55
V m 85*5
C a H 5 48-0
CN 30-2
O 7*4
2»Va 85-6
Vm 85*5 by formula
Digitized by v^ooQle
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS. 197
Phenyl carbimide C 7 H„NO
>.p. 163° d 50 1*092
C = 0*45 from
Vm 121*7
phenylthiocarbimide
c«h 5
92*8
CN
30*2
O
7*3
130*3
less for attach, to C 6 H,
-i *4
2»V a
128*9
Vm
1217
A
-7*2
Whilst ethyl carbimide is apparently normal, in that it pos¬
sesses the open-chain formula, and so all the alkyl carbimides,
phenyl carbimide, on the other hand, seems to possess abnormal
structure.
This is indicated by the large minus difference, and shows
that there is a three-membered ring in the molecule of phenyl
carbimide. Such a formula as that indicated in the preceding
scheme would be suitable. We must then suppose that phenyl
carbimide possesses the structure shown by the formula
C 6 H 5 —N — C >
\/
o
It will be seen that the other aromatic compounds of this
nature are quite normal.
TABLE
LXXII.
Compounds.
v m .
SnVa—R.
A.
b.p.
C.
C(jH B . c ; n . . .
121*6
123*0
-l* 4
—
—
C,H„ . N : C > . .
121*7
123*0
“ **3
—
—
C„H, . S . C : N . .
143*4
145*0
-1*6
I 3 i°
—
C,H t . N : C : S . .
143-6
145*0
- 1*4
219*8
0*45
Thus there is for the above compounds a contraction of - I *4,
owing to the attachment of the unsaturated groups to the nucleus.
The Amines.
The amines are derived from the cyanides or nitriles by the
addition of nitrogen.
Digitized by v^ooQle
198
LIQUID CHEMICAL COMPOUNDS.
There is in the process of reduction a remarkable contraction.
2 h 2 + c 2 h 8 . c : n c 2 h 5 ch 2 . nh 2
14-8 78-2 93*o
The real volume of propyl amine is
C 8 H 7 NH 2 Vm 85*2
A = 85*2 - 93*0 = - 7-8.
Reduction in the cyanides thus produces a contraction.
The starting-point of the amines is ammonia.
H
/
N—H
Vm
26*9
\
H
N
i 5‘8
3 H
ii*i
2nV a 26*9
V*, 26*9
This maximum value of nitrogen is preserved in such simple
compounds as NH S , NC 1 3 , N = N, etc., and in the complex
tertiary amines, as will be shown.
In considering the amines we are met with the feature of a
divided chain. This can be seen from the plane formulae
ch 3 ch 8 ch 8
ch 8
\ /
N
There are important points of difference, however, between
these two classes of compounds.
TABLE LXXIII.— The Paraffins.
b.p.
A.
V.
A.
ch 3 . ch 2 . ch 2 . ch 3
i°
-18
96*0
CH(CH s ) 8
-17
CHg . CH 2 . CH 2 . CH 2 . CH a
38
-8
117*8
-0*4
CH 8 . CH a . CH(CH 8 ) 2
30
-28
XI 7"4
C(CH 8 ) 4
10
ch 8 . ch 2 . ch 2 . ch 2 . ch 3
7 i
-9
139-9
-0*5
CH 8 . CH 2 . CH a (CH s ) a
62
“7
139-4
CHg. CH(C 2 H 5 ) 2
64
-13
- 3 *i
(CHg) a . CH . CH . (CHg) a
58
-28
136*8
- 1*0
CHg . CHg . C(CHg)g
43
-23
138*9
or 48
Digitized by v^ooQle
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS. T 9 g
It is seen that the boiling-point becomes depressed as the
hydrocarbon chain becomes more and more branched. The
volume is correspondingly diminished. Additional data are
however, urgently needed.
TABLE LXXIV.— The Amines.
b.p.
A.
V.
H a NC a H,
18 0
-10*8
64*6
HN(CHJ S
7*2
h 2 nc 3 h,
49 *o
“ 35*5
857
+ 2*3
N(CH 3 ) 3
3*5
88-o
h s nc 4 h.
76*0
— 20*0
106*8
+ 2*5
HN(C 2 H,) 3
56*0
109*3
H 2 N(C,H 13 )
130*0
-41*0
148*0
+ 4*6
N(C 2 H,) s
89*0
152*6
The results of the above table show, that, whilst the boiling-
points of the two series of compounds diminish as the chain
becomes branched, the series varies as regards volume. In the
paraffin series the volumes contract as branching occurs, in the
amines they expand as a result of this circumstance. In the
first series, the boiling-points and the volumes vary in the same
sense, among the amines in opposite senses. This at least
shows that the boiling-points are not always a guide as to the
manner in which the volumes will change. We now examine
the volumes in order to ascertain, if possible, the modes of
increase.
We have already made use of the principle of constant rela¬
tive volumes, for example, in discussing propyl and allyl alcohols.
We now apply it to a discussion of the amines. It may be
stated in passing that, until this principle was applied to this
group of compounds, the volume relations appeared very perplex¬
ing, and the results not very trustworthy. As we shall see by
its utilization, nearly all the difficulties vanish.
If we consider propyl and allyl amines, we find
CH S . CH a . CH 2 NH 2
CH a : CH . CH 2 NH 2
V m .
857
78*5
A for H 2 7*2 (2 x 3*6)
Digitized by v^ooQle
200
LIQUID CHEMICAL COMPOUNDS.
N(C 8 h 7 ),
N(CjHj),
222*53
20070
6H
21 83 (6 x 3*638)
1
N(CH 3 ),
88-o
N(C a H t ),
152*6
3CH a
64*6
CH,
21*6 (6 x 3 60)
We thus notice once again the familiar relations
CH 2 = 6H C = 4 H H = 3 *6o
Among the primary amines we also find that the following
similar value holds.
Vm.
C 2 H 8 NH 2 646
C 3 H 7 NH 2 857
CH 2 21*1
The relations already indicated will enable us to ascertain
the volume of nitrogen in the three series of amines.
CjHjNHj
CjHjNH,,
V m .
64*6
857
2»V a (R).
54 '°
75 '®
N.
io*6
10*1
b.p.
18
49
NH . CjH,
109*3
97-2
12*1
7*2
N(CH,),
N(C 2 H 6 ) s
88*o
152*6
75-6
r 4°'4
12*4
12*2
3*5
89
3CH 2
SSl
64*6
21-53
222*53
20070
207-36
i 85'54
15*17
15*16
156*5
155*5
6H
H
2X-83
3638
The values are apparently from 10-11 in the primary amines,
12-15 * n the secondary and tertiary; the volume of the nitrogen
atoms thus varies in volume in the different compounds.
There are many peculiarities connected with this series, and
one of the chief is the unexplained value for nitrogen. The
most useful point of view in which to regard the amines is to
compare them with the alcohols and ethers.
Digitized by
Google
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS.
201
Alcohols.
CH 3 . CH 2 . CH 2 . CH 2 . OH
Vm.
ioi*8
Primary amines.
CH 3 . CH 2 . CH 2 . CH 2 . nh 2
Vm.
l° 7 * 4
Secondary amines.
CH 3 . CH 2 . NH . CH 2 . CH 3
iog*3
The ethers.
Tertiary amines.
ch 3
CH 3 . CH 2 . CH 2 . O . ch 3
ch 3 . ch 2 . O . CH 2 . ch 3
105*0
io6*3
CH 3 . N . CH 2 . CH 3
CH 3 . N(CH 3 ) 2 + ch 2
iog*6
We notice that the amino group (NH 2 ) is similar to the
hydroxyl (OH) group, in that it occasions the smallest volume,
when in the terminal position. As oxygen becomes situated more
and more within the molecule, the volume increases, so with
nitrogen. Moreover there is an additional increase as nitrogen
becomes the centre for the attachment of three carbon atoms.
It becomes more and more evident, as we study the subject,
that we must suppose the typical atoms O, N, S, etc., to produce
specific effects * on the hydrocarbon radicals and other groups
with which they are combined. These result in modifications of
the individual atomic volumes of such groups, whilst as a rule
the relative atomic volumes are preserved. Thus we have shown
that the value V/W measures the volumes of the hydrogen
atoms or equivalents in the different series. If now we consider
similar terms of the different series
ch 3 . ch 2 . ch 2 . CH 3
ch 3 . ch 2 . ch 2 . CH 2 C 1
ch 3 . ch 2 . CH 2 . CH 2 I
CH S . CH a . CH 2 . CH 2 OH
ch 3 . CH 2 . CH 2 . CH 2 SH
we shall find different values of V/W. Morever, the shape of the
different curves may not be quite the same, or they will differ in
some particular.
If we take the normal paraffin series as the standard, and
consider the value of the radicals CH 3 , C 2 H 6 , C 8 H 7 derived
therefrom as the reference values, we shall, on considering other
series, sometimes find that those which we have to attribute
to these groups are different. It may also be true that only
certain parts of molecules are affected by constitutive influences.
In any case, we must consider that variations from the normal
are due to the specific influences of certain atoms or groups, as
* These influences are not necessarily direct, but in any case the atoms N or O
or the reactive groups containing them are the causes of such effects.
CH 3 . CH 2 . O . CH 2 . CH 3
CH 3 . CH 2 . S . CH 2 . CH 3
CH 3 . CH 2 . NH . CH 2 . CH 3
Digitized by v^ooQle
202
LIQUID CHEMICAL COMPOUNDS.
well as to differences of arrangement. The constitutive influences
in question will affect the volumes and probably the boiling-
points. These definite statements are the outcome of the work
done on the nitrogen compounds, especially the amines, and are
considered to be of the first importance.
It is seen that the primary amines are considerably diminished
in volume as compared with the corresponding secondary and
tertiary compounds. In this they are analogous to the alcohols
as compared with the alkyl ethers.
As we have seen, there is a contraction of -7*5 approximately,
when the cyanides are transformed into the amines by reduction.
Similarly, there are contractions of - I *5 to - 3*5 when the
aldehydes and ketones are transformed into primary and second¬
ary alcohols by reduction.
We have considered these diminutions in volume as involving
only the nitrogen or oxygen atoms, but it is possible that all the
atoms in the groups
—OH and —NH a
may be concerned. In any case the constitutive effects are prob¬
ably due to the oxygen and nitrogen atoms respectively.
The alcohols and the primary amines may also be shown
to be similar in many other respects, but it may be noted that the
low relative value of the primary amines and the alcohol is con¬
nected with the terminal position of the reactive groups NH 2 and
OH.
It is to be remarked that the atomic volumes of nitrogen and
oxygen both reach a maximum in the simplest compounds,
become depressed as the complexity increases, and finally reaches
about the maximum value in complex symmetrical amines and
ethers respectively.
Vm .
0 .
Vm.
N.
H—O—H
18-5
II*I
NH,
26*9
15*8
CH S —O—CH,
62*1
10*1
N(CH 8 ) s
N(C 8 H 7 ) s
88*o
12*4
Cj|H 7 —0—C 3 H 7
151*3
n *4
222*5
15*2
c 4 h„—o-c 4 h„
I 97 '®
n *3
The volumes of amines with
iso-groups.
Vm.
2«Va(R) - 0*5.
A.
b.p.
(CH s ) 2 CH . 1
ch 2 nh 2
106*3
103*7
-2*3
68°
(CH s ) a . CH . CH a
. ch 2 nh 2
127*0
129*4
-2*4
95 °
Digitized by v^ooQle
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS. 203
There is in this class of compounds a depression of about
-24.
Compounds Allied to the Amines.
Ethylene diamine.
NH 2 . CH 2 . CH 2 . NH 2 b.p. 116*5° d 0 0*902
Vm 75 *o
2CH 2 43*2 (12 x 3*6)
2NH 2 35*6 ( 2 x 17*8)
SnVa 7^*8
Vm 75*0 by formula
A -3*8
The large contraction which has been associated with a /3
compounds is here again evident.
The formula is
CH 2 -NH >v
1 '
CH 2 -NH^
or
and the contraction is an expression of the effect due to the con¬
tiguity of the NH 2 groups.
Propylene diamine C S H 6 (NH 2 ) 2
b.p. 120°
d 1B 0*878
C = 0*50
CHo-CH-
Vm 95*5
1
c 3 h 6
64*8 (3 X 21*6)
1
j
2NH 2
35*6 (2 x 17*8]
CEL -
-Nil'
2
a
2nVa
100*4
Vm
95*5
A
-4*9
Digitized by v^ooQle
204
LIQUID CHEMICAL COMPOUNDS,
Pentylene diamine C 5 H 10 (NH 2 ) 2
CH,—CH,—NH.
b.p. 179 0
d# 0-9174
/
ch 2
Vm
134*4
\
ch 2 —ch 2 —nh 2
c 5 h 10
io8*o
2NH 2
35-6
2nVa
143-6
Vm
* 34*4
A
- 9*2
From these calculations we see that there are large contrac¬
tions connected with a/8, ay, aS, etc., compounds. These have
been considered in the main to be due to a curvature of the hydro¬
carbon chains, owing to the close approximation of the two
amino groups.
That this is so, is shown by a possibility of the formation of
ring compounds from the diamines. For instance, by the splitting
off of ammonia, the rings become closed and the cyclic alkylene
imides are formed.
CH a — ch 2 —nh 2 ch 2 —ch 2
/ / \
CH 2 -> CH 2 NH + NH S
\ \ /
ch 2 —ch 2 —nh 2 ch 2 —ch 2
It is doubtful, however, if the whole of the contractions
shown above are due to a curvature of the chain.
Dimethylene compounds
CH,X,
A= - 3 ’i
Trimethylene „
CjHjXj
A = - 4*5
Then in pentamethylene ,,
' CjHjdX,'
A = -( 3 -I + 3 x ** 5 )
= - 7*5
The following is a study of the ring compounds pyrrolidene
C 4 H 9 N and pipiridene C 5 H n N, which are derived from the
diamines.
Tetramethylene diamine
and pentamethylene diamine
Pyrrolidene C 4 H 9 N
CH,-CH 2 b.p. 86-8
\
^,NH
C 4 H 8 (NH 2 ) 2
C b H 10 (NH 2 ) 2
do 0-879
Vm 90*0
C = 0*48
CH*
-CH a
We now compare the contractions ob¬
tained on the basis of the values H = 3*7 and 3*6 respectively.
Digitized by v^ooQle
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS. 205
4 CH a
88*4 (4 x 22*1)
4 CH a
86*4 (4 X 21 * 6 )
N
12*0
N
12*0
H
3*6
H
3-6
104*0
2 nVa
102*0
Vm
90*0
Vm
90*0
A for O5
-14*0
A for Os
- 12*0
Pipiridene C 5 H n N
V*» 108*8 (Schiff)
sC $
108*0
12*0
CH 2 -ch 2
/ \
H
36
CH 2 NH
'XtN a
123*6
\ /
Vm
108*8
CH 2 -CH 2
A for Oe
-14*8
The calculation for pyrrolidene which shows a value of 12 is
the correct one, since it contains a five-membered ring. From
this we see that in the diamines the volume of CH 2 = 21*6 and
not 22*1.
Thiophene C 4 H 4 S Benzene C 6 H 6
CH
CH=CH
/ %
CH CH
CH
CH CH
\ ✓
CH
V m 85*0
(Schiff).
Vm
96*0
4CH
74-0 (4 x 18-5)
6CH
111*0
S
22*0
*nV a
96*0
Vm
85*0
Vm
96*0
A for Os
- 11*0
A for Oe
-15*0
These numbers agree with those found on the basis of the
diminished values of V/W or the hydrogen equivalent. This
rule apparently does not apply to the unsaturated ring com¬
pounds.
Pyrrol C 4 H 5 N
CH-CH
CH CH
\ /
NH
b.p. 131 0
dws 0*9752
C = 0*46
Pyridene C 5 H 5 N
CH
^ \
CH CH
CH CH
\ /
N
Digitized by v^ooQle
206 LIQUID CHEMICAL COMPOUNDS.
V* 79-4
Vm io8‘8
(Schiff)
C 4 H 4 750
c 5 h #
92*5
H
36
N 12*0
N
12*0
90*6
104*5
for Os -n*o
less for 0«
-150
2 hV u 79*6
2*V a
89*5
V» 79-4
Vm
-a
89*5
Methyl glyoxylin
c 4 h,n 2
C
:h (
:h
b.p. 199 0 dj.
1-0363
C = 0-46
l
H
V*.
93*8
\ /
C.H.
81-4
c
1
N
24-0
ch 8
105*4
less for Os
-11*0
94*4
V w 93*8
CH
CH
Quinolene C 8 H 7 N
CH CH
CH C CH
^ \ / \
CH C CH
I 11 1
1 11 1
CH C CH
CH Cl CH
\ /\ ✓
\ /’\ S
CH CH
CH N
Vm 147*2
Vm 140*0
Ci 0 H 8 177*6 (48 x 3-7)
C 8 H 7 159-1
N 12-0
less for ring - 30-4
2»V a 147*2
171*1
V m 147*2
less for ring - 30*4
2nV a 140*7
V m 140-0
A few additional ring compounds will be dealt with at the
end of the present chapter.
It will be observed that we have found a value for nitrogen
similar to those derived from the secondary and tertiary amines.
This is analogous to the results obtained for rings including
oxygen and the volume of this atom was found to be similar to
the values derived from the ethers. We find in this fact one
more analogy between the amines and the ethers.
We now turn our attention to another class of compounds,
the carbylamines, the isocyanates, and the thiocarbimides.
Digitized by v^ooQle
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS . 207
It will be noticed that they all have one feature in common.
Carbylamines R—N = C >
Isocyanates R—N = C = O
Thiocarbimides R—N = C = S
This feature is, that the alkyl groups are in every case attached
to nitrogen. They may thus be regarded as derivatives of the
amines. As such, the alkyl groups are likely to suffer the con¬
tractions already noticed, these being due to the interaction with
the atom nitrogen, that is, CH 2 = 21-6 and not 22*i.
In order to facilitate comparisons, we give in the following
table, not only members of the series just indicated, but also
those of the isomeric compounds in which the alkyl groups are
attached to the carbon or sulphur atoms.
TABLE LXXV.
Cyanides.
R.C : N
b.p.
Vm.
A.
V*.
b.p.
Carbylamines.
R—N : C >
C 2 H 5 . C : N
0
o>
78*4
- 1*2
77*2
79 *o°
C 2 H 5 . N : C >
Cyanates
or cyanetholens.
R.O.C ; N
C 2 H 5 .0 . C : N
0
0
5 »Va.
86*9
-i *3
v w .
85*6
CT>
O
0
Isocyanates.
R . N : C : 0
C 2 H 5 . N : C : 0
Thiocyanates
or sulphocyanides.
R.S.C : N
C,H fi . S . C : N
146°
V m .
100*1
-o*8
Vm.
99*3
* 33 °
Thiocarbimides.
R . N : C : S
C 2 H fi . N : C : S
C 8 H 5 . S . C i N
161 0
115*0
-i *7
ii 3*3
150*7°
C s H fi . N : C : S
C 5 H u . S . C : N
—
166*5
-27
163*8
182*0°
C 5 H n . N : C : S
It is seen that the compounds which possess the
R—N =
group have lower boiling-points, and consistently smaller
volumes. Contractions are thus found which increase as the
alkyl groups become more complex. As before indicated, this is
probably due to specific actions of the nitrogen atoms on the
radicals. The validity of this rule is shown in the following table.
TABLE LXXVI.
H 3-6 CH 2 21*6
Compound.
Vm.
N : C >
+ R
5 nVa.
C 2 H b . N : C >
77*2
30*2
N : C : 0
46*8
77 *o
C 2 H 6 . N : C : O
85*6*
37*6
N : C : S
»»
84*4
C 2 H 5 . N : C : S
99*3
52*4
»»
99*2
C 8 H 5 . N : C : S
H 3*3
»>
6 i*2
113*6
C 6 H u . N : C : S
163*8
»»
111*6
164*0
* The volume of this compound agrees better with the usual value of C 2 H 5 48
S»V a 48*0 + 37*6 = 85*6.
Digitized by v^ooQle
208
LIQUID CHEMICAL COMPOUNDS.
Aromatic Amino Compounds.
The interesting subject of the possible mutual influence of
the phenyl (C 6 H 6 ) and amino groups is capable of fairly extended
treatment, owing to the extended material which is available.
It will be necessary to give detailed calculations in order to illus¬
trate the methods of calculation. The calculation of values by
the direct summation of the individual atomic volumes is not
advisable when compounds of some complexity are dealt with,
but comparisons with closely analogous compounds are pre¬
ferable.
We first compare aniline and picolene, both of which possess
the empirical formula C 6 H 7 N, but which possess somewhat
different molecular volumes.
Aniline C 6 H R NH 2
a Picolene
V m io6*6
V m 111*7
CH CH
CH
✓ \
CH CH
C
H CH
^c^nh 2
CH C-
\ /
N
-CH,
C 6 H 5 928
88*8
NH a 17-8
26*0
iio*6
N
12*0
less for const. - 4*0
126*8
less for £7«
-15*0
io6*6
--
Vm 106 *6
in-8
'
Vm
iii*7
It is seen that there is a contraction of - 4 0 connected
with aniline. This is in accord with the rule that when an
unsaturated group is attached to the phenyl radical, there are
contractions of greater or lesser magnitude. We now investigate
a number of compounds similar to aniline, but containing hydro¬
carbon radicals of different magnitudes.
Digitized by v^ooQle
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS. 209
p Toluidine
C 6 H 4 (CH 3 )(NH 2 )
b.p. 201*5 V m 128*9
(Neubeck)
CgH 4 89*6
CH 8 25*5
NH 2 17*8
132*9
less for const. - 4*0
2 nVa 128*9
Vm 128*9
Benzylamine
C 6 H 4 CH 2 NH 2
b.p. 183 d 14 o*ggo
V m 127*4
CH
✓ \
CH CH
CH CH
\ /
C-CHjNHj
l’S:
92*8
21*6
nh 2
17*8
132*2
less for const.
-4*0
2»Va
128*2
V m
127*4
The boiling-point of Neubeck’s compound has been taken
as 201*5°, but at 198° (the boiling-point usually given), benzyl-
amine has the same volume as p toluidine.
CH
✓ \
CH CH
CH C—NH—NH«
\ /
CH
d 22 1 *091 b.p. 241 0 C *a 0*45
Vm 120*0
c 6 h 6
92*8
NH
14*2
nh 2
17*9
124*9
less for const.
-4*0
2 *V a
120*9
Vm 120*0
From this we conclude that the contractions due to the com¬
bination of - CH 2 NH 2 with the phenyl radical QH 5 is about the
same in value as for NH 2 .
14
Digitized by v^ooQle
210
LIQUID CHEMICAL COMPOUNDS,
Dimethyl aniline.
\ m 152*4
,CH
CH
CH
CH
CH / CH 3
C^-N v
-..-'V
n ch
C,H„
N(CH,) t
less for const.
2nVa
3
92*8
62*8
Ethyl aniline.
HC 2 H
151*6
Vi
CH
/• \
CH CH
CH CH
\ /
C-NH(C 2 H 5 )
155-6
-4-0
151-6
151*6 ethyl aniline
152*4 dimethyl aniline
N(C cife
N(CHj),
88*o
25*2
62*8
Dipropyl aniline.
/C,H,
c,H 5 r /
\c,h 7
N(C c H fl >
C 8 H 7 08 3
N(C s H 7 ) 2 154*2
CjH 5 92*8
N (CjH,), X 54-2
less for const. - 4*0
Diallyl aniline.
C.H*
/^J n 5
C ‘ H ‘ N \c,h
N(C 8 H 6 ) # 200*7
C.H. 61*2
C 6 H 5 92*8
N(C 8 H 6 ) 2 1395
2470 N(C 8 H 6 ) 2 139*5
232*3
less for const. - 4*0
S»Va 243*0
Vf» 243*6
2»V a 228*3
V« 2257
- 2*6
Di-isopropyl aniline.
C 6 H 5 N{CH(CH 8 ) 2 } 2
V#» 235*9
Di-propyl aniline C 6 H B N(C 8 H 7 ) a 243*6
Di-isopropyl aniline C 6 H 6 N(C 8 H 7 ) 2 235*9
A -77
It is concluded from the above investigation that there is a
contraction of about - 4*0, when the amino or similar radical
becomes attached to the benzene nucleus. Moreover it does not
apparently matter what the complexity of the substituting hydro¬
carbon group or groups is, so long as they are saturated. If
Digitized by v^ooQle
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS. 211
unsaturated, in the sense of possessing residual affinity, it seems
that the contractions are increased. On this last point, however,
we need further information.
One very interesting point is, that if methylene groups be
interposed between the unsaturated radical and the nucleus, it
makes no difference to the contraction. We have a sufficient
number of instances of this to be certain on this point.
p Cresol.
Vtn.
Vm-SnV*.
V*.
Benzyl alcohol.
C 6 H 5 (CH s )(OH)
1237
“i *3
123*7
C 6 H 8 CH 2 OH
C 6 H 5 . O . CHg
125*2
Benzyl chloride.
2 nVa
134-9
-1*6
i 36*5
C 6 H 5 CH 2 C 1
not
Benzilidene chloride.
5 »V a
154*5
—
154*7
C 8 H 5 CHC 1 2
Benzpic acid + 2CH 2
Phenyl propionic acid.
C 6 H 5 . COOH + 2CH.
170*9
“ 3*8
170*9
C 6 H 5 . CH 2 . CH 2 . COOH
p Toluidene.
Benzylamine.
C«H 4 (CHj)NH 2
128*4
-4*0
127*9
C 8 H 5 CH 2 NH 2
It is difficult to understand why this should be so, unless by
a suitable arrangement of the atoms the phenyl and unsaturated
groups are brought into proximity. This is probable. Since the
amount of the contraction apparently bears some relation to the
reactivity of the groups or the number of unsaturated atoms, we
must connect these contractions with the residual affinity associated
with the unsaturated atoms, that is, with the latent valencies.
We now discuss a number of cases showing interference be¬
tween contiguous groups.
p Toluidene.
c^-ch 3
CH
CH
CH CH
"C^NH Q
b.p.
198°
V» i28*g
2 n V a 128*9
The Toluidenes.
m Toluidene.
*ch
CH
X
-CH.
CH
_ CH
^ C -^ H 2
b.p.
199 0
i*i . 127*8
A -2*2
between 0 and p.
14 *
0 Toluidene.
,CH
CH
CH
X
X
c-
-CH
A
C-NH
2
CH"
b.p.
197 0
126*7
Digitized by v^ooQle
212
LIQUID CHEMICAL COMPOUNDS .
The ethyl amido benzenes.
para.
CH
ortho.
%
c 2 h 6
*C ET¬
CH
■C 2 H 6 - v
CH CH
NH 0
b.p. 2i6°
Vm I49 *o #
b.p. 2i6°
147*7
A -1*3
153*9
C,H 4
89*6
less for const.
-40
C s H 6
4^5
less for 0 struct.
149*9
-2*2
NHj
17*8
153*9
2nV«
v w
't-
H M
The Xylidenes.
1 :
3 : 4
1 : 3
: 5
CH CH
CH X C
V/
ch 3
,c^ch 3
CH CH
-CH 3 - n
\
NH 2 "'
b.p.
218°
V_x
Vm 148*6 (Neubeck)
ch 3 —
b.p.
220
c— nh 2
CH
d 16 0*972 C = 0*52
CeH,
2CH,
NH a
less for const
less for 0 struct.
2 nVa
Vm
86*4
51*0
17*8
155*2
- 4 *o
151*2
- 2*2
I 4 9*0
148*6
Vm 151*2
NH»
less for const.
2 nVa
Vm
86*4
5 i*o
17*8
155*2
- 4 *o
151*2
151*2
* The m and even p compounds probably suffer appreciable contractions.
Digitized by v^ooQle
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS. 213
The Oxides and Esters of the Nitrogen Acids.
Nitrous oxides.
N 2 0 .
This compound boils at the comparatively low temperature
of 186° A.
Its volume at the boiling-point can be calculated by D.
Berthelot’s formula from the critical data.
M
D
Vm
ii*iT 2 c
P c (2T c - T)’
P c , T c crit press, and temperature. T boiling-point temp.
Compound.
T.
D.
T k .
P K (atmos.) V w =
V. calc.
n 2
78*6
0*885
127
33 *o
31*6
30*9
NH,
274*0
0*635
403
115*0
26*9
29*5
n 2 o
183*0
—
309
73*6
33 *i
Vm for N a O is 33*1
2N 31*6 (2 x 15*8)
O 8*3
2nV a
Vm
39*9
33-i
A = -6*8
This large difference between the theoretical and experi¬
mental values points to ring structure. On considering the type
of ring which would be suitable, we find that it is compatible with
a three-membered single ring. Thus
<N
N>
/
<N
This is indeed the only type of structure which is in agree¬
ment with our conception of the valency relations of the atoms
nitrogen and oxygen. It is the only one which would take into
account the well-known trivalent nature of the one atom and
divalency of the other.
This result is of extreme importance in accounting for the
constitution of the phosphorus oxides.
Nitric oxide
NO
Digitized by v^ooQle
214 LIQUID CHEMICAL COMPOUNDS,
The critical data for this compound are
T c i8o°A P c 71 atmos. Tb.p. i20°A.
By D. Berthelot’s formula
v~ = M _ ii*i x 180 x 180 359600 _ ot . o
m D = 71 X (360 - 120) ” 17040 = 1
The theoretical volume would be
N 15*6
O 8-2
2nV a 23‘8
The difference is not too great for this kind of calculation.
Nitrosyl chloride.
O = N - Cl
The data for this compound are
d-15 1*425 b.p. -5*0° C = 0*46
Vm 46*7
N
15*6
O
8*3
Cl
22*1
2 w V a
46-0
Vm
467
Theory and experiment are seen to agree quite well, and
also that the divergence from the theoretical value noted in the
case of nitric oxide NO, is due to the particular method of
calculating it.
We are also certain that the true volume of NO is
NOCl 467
Cl 22*1
NO 24*6
5»V a 23*8
Moreover the formula for nitrosyl chloride cannot be any
other than
o = N—Cl.
Nitrogen trioxide.
n,o 8
Digitized by v^ooQle
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS. 215
Data:—
d 0 1*449 b.p. 4 0
d 4 1*437
V» 53*9
N a 31*6 (2 x 15*8)
3O 24*9 (3 x 8*3)
2«V a 56*5
Vw 53*9
A -2*6
It is seen that a difference of 2*9 exists between theory and
experiment This is not sufficient for us to suppose that there
is any great modification in structure from the usual one, but in
attempting to find an explanation we notice one remarkable
fact, that the sum of the volumes of N 0 2 and NO are almost
exactly equal to the experimentally derived volume. In the
latter case, however, we must use the volume found by D.
Berthelot’s formula and not the one agreeing with theory.
NOn
NO
32*0 ( v . ante)
23*8
2»Va 54*8
Vm 53*6
This would lead us to suppose that the liquid at 0°, or only
four degrees below the point at which it vaporizes and is sup¬
posed to decompose, is really a mixture of the two liquids NO
and N0 2 , a supposition which may quite well turn out to be true.
The formation and decomposition would thus be
O
N > + O N = O
/
—>
\
\
N >+ 0
N
N = O
/ v
H-o
O = N = O
V
& = 0
Possibly all of the above compounds may exist at some
temperature or another, and no doubt together in greater or less
proportions.
Nitrogen tetroxide
N 2 0 4
This compound has been actually examined by Thorpe at the
boiling-point, and the volume found by him was
Digitized by v^ooQle
2l6
LIQUID CHEMICAL COMPOUNDS .
Vm 64*0 very approx. (63*95 exactly) at the boiling-point 21*6
NL 31*2 (2 x 15*6)
0 33*2 (4 x 8*3)
64*4
V» 64*0
In considering the question of structure we might be led to
suppose that its constitution is represented by
o o
II II
N-N
This is especially the case if we note that the volume of the
nitro group in the nitro paraffins is half the volume of N 2 0 4 or
320.
C ’ H $:
80*4
48*0
NO,
32*4
Certain observations like the following have led chemists to
suppose that it has a different formula.
2 NO } 0 + H ,0 = O + 2NO a . OH
This is indicated by the one just given, which is—
O
ll
N—O—N « O
II
O
or a mixture of the oxides of nitrous and nitric acid. This is
quite compatible with the molecular volume found by Thorpe
for NO 24*1 from NOC 1 .
NO« 32*0 from value in nitro paraffins.
O 8*3
2 tNa 64*4
V m 64*0
The nitro paraffins.
V m . 2»V a .
CH S . NO a 59*6 58*0
Digitized by v^ooQle
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS . 217
The volume of the methyl compound in any series is usually
larger than that calculated.
c 2 h 8 no 2 v 80-4.
This has already been calculated
C 2 h 6 48*0
NO a 32*0
%Na 8o*o
V m 80-4
The general formula applicable to these compounds is
O
/
not K—N
The esters of nitrous acid.
Vm* b.p.
ethyl nitrite C 2 H 8 —O—N = O 7g*2 + 16 0
If we suppose that the oxygen atom in the C 2 H 6 —O— group
possesses a similar volume to that found in the esters of the
carboxylic acids, we obtain the following results :—
C 2 H 5 48 *o
o 7-4
NO 23*9
79*3
Vm 79*2
The formula is that already given.
R—O—N = O
It would be advantageous to possess a larger series of values.
An examination and comparison of the boiling-points shows
how large a difference exists between the isomers in this respect.
Ethyl nitrite. b.p. nitro ethane. b.p.
C 2 H 5 —O—N = O 16 0 C 2 H 8 —N0 2 114-5
There is no less a difference than 100 degrees between the
boiling-points. In spite of this fact the law of additivity holds.
There is also the great probability of the nitro paraffins being
associated.
O
II
R—N
II
O
Digitized by v^ooQle
2 iS LIQUID CHEMICAL COMPOUNDS.
The formation of nitrous acid would be represented thus
H O =
H O
H O
>N
H
c
I— O
H O
H O
H O,
H O
V
\ *
-OH
-OH
H 0 = N/ OH
Water then splits off
H --^N—OH = HO—N = O
h\o/
We have no knowledge of the volume of nitrogen pentoxide
but its formula is doubtless
o = N = o
i
O = N = O
United with water this gives
Nitric acid HNO s .
d 0 i*54 b.p. 86. C = 45°-o
V« 45*o
N0 2 32*0
HO n*o
5nV a 43*o
Vw 45*0
The excess of V TO over 2 n V a is due to the unusually large
volume of hydrogen which is found in such compounds. Compare
HCN! HCl etc.
The esters of nitric acid.
Vm. b.p.
Amyl nitrate C B H n —O—N0 2 152*9 (Schiff) 147 0
c 5 h u
114*1
0
7*4
no 2
32*0
2nVa
153*5
Vm
152*9
Again the volume of the oxygen atom in
R—o
is found to be 7 4 as in the carboxylic esters and acids,
Digitized by
Google
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS . 219
The question now arises whether the constitution of the
remaining group NO a is—
o
II
—N or —0—N = O
II
O
Presumably the formula suited to the nitrates is—
o
II
R—O—N
II
O
but there is difference of opinion with reference to nitric acid.
Bruhl, 54 from optical data, finds that nitric acid may have a
peroxide formula
H—o—O—N = o
but supposes that the formula represents only one phase of a
cycle of changes, of which the formula H—O—NO g may be
another. Klason and Carlson apparently confirm this view by
the detection of the alkyl peroxide in the products of saponifica¬
tion of the alkyl nitrates.
From our present point of view, we may say that there is
little or nothing in the way of evidence either for or against this
view, since it is probable that the volume of HNO s obtained
would fit either formula. The reason is that the volume of
the singly bound oxygen attached to nitrogen may not be very
different from 8*3, the volume of: O.
The Aromatic Nitro Compounds.
It has been shown that when unsaturated groups are attached
to the benzene nucleus, there are considerable contractions.
We have seen that the group N 0 2 possesses a volume of
32*0 units when attached to the alkyl radicals and also in N 2 0 4 .
O
✓
and O = N—O—N
\
If, however, the hydrogen atoms in the methyl group be
substituted by chlorine atoms, the conditions are realized for that
Digitized by v^ooQle
220 LIQUID CHEMICAL COMPOUNDS.
interaction between the two groups which might be responsible
for a contraction.
Chloropicrin.
Cl O
II
CHCl,
84*5
Cl C N
i, 1
Vm 110*5 (Thorpe).
CC1,
8o*8
H
37
no 2
32*0
CC1 3
8o*8
S*Va
112*8
Vm
110*5
A
-2*3
The operating influences are probably derived from the
nitrogen atom rather than from the doubly-bound oxygen.
Contractions like the above are shown when the nitro group
is united to the benzene nucleus.
Nitro benzene C ft H s NO a
b.p. 208° db.p. 1*0073
V m 122*1
^ CH \
CH CH
CH CH
CH
cn
b.p. para
144-8
238 '-
2»Va 144-4
C,H, 928
NO, 32*0
2nV a 124*8
Vm 1221
A - 2*7 for constitution.
The nitro toluenes.
,CH
CH
<?H
\
CH
•z:,
•CH,
N0 o
meta
144-6
.CH
CH
\
C-CH
3v
H 0.—NO,
.
ortho
143*0
-i*8
Digitized by v^ooQle
MOLECULAR VOLUMES OF NITROGEN COMPOUNDS. 221
Orthonitro toluene.
c 6 h 4
89*6
no 2
32*0
ch 8
25*8
147*1
less for N 0 2
- 2*7
144*4 • • • volume of paramtro toluene
less for 0 struct.
-1*8
2wV fl
142*6 • . . volume of orthonitro toluene
Vm
I 43 *°
1
Nitro metaxylene C 6 H 8 (CH 8 ) 2 N 0 2 (1:3:4)
b.p. 244 0 db.p. 0*9163
V m 164*8
CH
OH
-CH„
X 3
CH
C — CH,
no 2
C 6 H 8
2 dH 8
86*4
5 i*o
no 2
32*0
169-4
less for N0 2
-27
166*7
less for 0 struct.
-i*8
2nVa
164-9
Vm
164*8
Nitroortho xylene C 6 H 8 (CH 8 ) 2 N 0 2 (1:2:3)
b.p. 252 db.p. 0*926 Vm 163*0
c«h 3
*CHj
NO,
less for N 0 2
less for 0 struct.
less for o struct.
Va
Vm
86*4
51*0
32*0
169*4
169*0
-27
166*3
-i*8
164*5
-i*8
162*7
163*0
One or two more ring compounds may now be studied. They
are interesting owing to the fact that they are either alkaloids
or connected with alkaloids.
Digitized by v^ooQle
222
LIQUID CHEMICAL COMPOUNDS ,.
CH 2
/ \
CH CH«
For propyl benzene
C 8 H 6 . C s H 7 C ss 0*50
Coniine
d 1B 0*850 b.p. 167°
V m 175*6 by formula
The most direct method of calculation is as follows:—
t„ i
H--CH,—CH,—CH,
Pipiridene
C *
2«V«
Vm
108*8
66*3
175*1
1756
Nicotine c,„h m n s
N
^ \
CH CH N(CH,)
I II / \
CH C—CH CH,
\ /
CH
—vuj
Jh,—Ah,
b.p. 247 0 d 4 1-033
d 15 i*oi 1 C =» 0*46
V m 193*1 by formula
Pyridene 89*3
Methyl pyrrolidin 110*4
199*7
less 2H - 6*4
2nV a 193*3
Vm I93*i
Tropilidene C 8 H 18 N
do 0*9665
b.p. 162° C
= 0*46
CH,-CH—
—c
:h 2
Vm 149
Ah
c 8 h 13
166*5
|
N
12*0
CH
1
178-5
k(CH a )-CH—
—1
k
less for ring
-30-0
InVa
1485
Vm
149-0
The above remarkable series of ring compounds indicates
how closely the theory of molecular volumes is able to synthesize
the values found by experiment or by calculation by an independ¬
ent method.
Digitized by v^ooQle
CHAPTER VII.
THE MOLECULAR VOLUMES OF PHOSPHORUS COMPOUNDS.
The Element Phosphorus.
The element has a very much smaller volume in the free than in
the combined state.
P = 20*9 (Ramsay and Masson) free
P = 27*0 combined
be
The complexity of the phosphorus molecule is well known to
P 4 .
It follows that the molecular volumes are—
P 4 (free) 83*6
Equivalent P 4 (combined) 108*0
A - 24*4
This is a. large contraction which can only be accounted for
by supposing that the phosphorus molecule involves ring struc¬
ture of a complicated kind. This is difficult to understand, if we
consider the nature of this molecule. The simplest assumption,
consistent with the valency relations of the phosphorus atom is—
P-P
>-P
but this would account for a contraction of only -8*5. A modi¬
fication of the above
P-p
P-p
which is equivalent to four three-membered single rings is found
223
Digitized by v^ooQle
224 LIQUID CHEMICAL COMPOUNDS .
to be suitable. Such a configuration would possess a contrac¬
tion of
4 x -6-o = -24*0
and this is practically the same as the number just found.
The above formula can be modified in a very interesting and
suggestive manner, which is shown in the following diagram.
This represents a regular tetrahedron, with an atom of
phosphorus at each comer, and the valency linkages directed
along the edges of this tetrahedron.
Such a formula explains many facts connected with the com¬
binations of this atom, and also of that of the very similar
nitrogen.
The valency directions of the single atom of either element
would be as in the diagram
N itrogen. Phosphorus.
and the formulae for the chlorides, etc., represented by
From the point of view of their spatial relations, these
formulae at least allow of union with additional atoms, to form
compounds which involve pentavalent nitrogen or phosphorus.
Digitized by v^ooQle
VOLUMES OF PHOSPHORUS COMPOUNDS.
225
The molecule nitrogen would be represented by
or by a modified formula.
In giving the configurations of certain simple phosphorus
combinations, we assume that a molecule of phosphorus P 2 (like
N 2 ) is at least theoretically possible, and rely on certain observa¬
tions on the molecular volumes of nitrogen compounds, and on
the recognized empirical formulae of others, for evidence concern¬
ing the genetic relations of the series of phosphorus compounds.
N
N
Elementary
nitrogen
P P
I +
Simple phosphorus
(not known)
N
\
0
N
Hyponitrous
oxide
+
'o
Phosphorus
sub-oxide
N = 0 N = 0
i\ v \
O or o
n<=o n=o
Nitrogen trioxide.
15
Digitized by
Google
226
LIQUID CHEMICAL COMPOUNDS.
Phosphorus oxide
The formulae for hyponitrous oxide and nitrogen trioxide
have been verified by means of molecular volumes. We see
from the above formulae that similar simple compounds occur
for phosphorus as for nitrogen but by simple combinations the
more complex compounds are found to be possible. The
molecular volume of phosphorus oxide might easily be found
experimentally. Failing direct data, we calculate the volume of
this compound by means of the formula already used.
O
o
4P
108*0 (4 x 27*0)
PCI,
93*34
60
48*0 (6 x 8*o)
P 0 C 1 ,
ioi*37
156*0
0=
8*o
less for O4
-8*5
147*5
less for 2 /~l ?
“i 3 *o
2«Va
134*5
dj 5 1*936 b.p. 173*1° C = 0*50
V OT 132*5
The difference between the two numbers V m and 2 n Va is
comparatively small. At any rate, it is satisfactory to find that
there is ample evidence for a large contraction due to more or
less complicated ring structure. (A = - 23*5 as against - 21*5.)
Thorpe has determined the volumes of a number of phosphorus
compounds experimentally, and this not only enables us to find
the atomic volumes, but also to give the probable structural
formulae of the different compounds.
Digitized by v^ooQle
VOLUMES OF PHOSPHORUS COMPOUNDS.
227
Cl
1
Phosphorus trichloride.
PC1 8
' Vm 93*3
<P—Cl
1
P
27*0
Cl
3C1
66 *o
2nVa
Vm
93*o
93*3
Cl
1
Phosphorus oxychloride.
POCI 3 Vm 10137
u
J-
II
PCI,
93*o
Cl
0
8*3
2nV a
Vm
101*3
101*4
Cl
:-L
i,
Cl
Phosphorus sulphochloride.
PSC1 3 V m n6*i
P 27*0
CL
66*o
2 „V a
Vm
115*0
ii6*i
Phenyl phosphochloride.
C.H^Clg
X
C6H5-
Br
■P—Br
L
Cl
1
V m 1617
. 1
-p>
1
c.h 5
PC 1 2
92*8
71*0
1
Cl
less for const.
163*8
— 2*0
2«V a
161*8
Vm
1617
Phosphorus tri-bromide.
PBr s
Bromine appears to have a similar volume
to that of phosphorus.
Vm 108 *3
108*0 (4 x 27*0)
IS *
Digitized by v^ooQle
228 LIQUID CHEMICAL COMPOUNDS.
Prideaux has given the volume of PC 1 6 and PBr 5
Phosphorus pentachloride.
PC1 8 Vm 128*9
PC1 5 128*9
PCI, 93'3
2C1 35-6
Cl 17-8
Phosphorus pentabromide.
PBr, V m 157-3
PBr, i57'3
PBr 3 108*3
2Br 49*0
Br 245
The usual volumes of the two atoms in question are
Cl 21*6 and Br 27*0
so that there are contractions amounting to
Aj = 17*8-21*6 = - 3*8^
[per additional atom.
A* = 24*5 “27*o - - 3 * 5 J
These are nearly equal values and they represent total con¬
tractions of
-7*6 and -7*0 respectively.
If the formulae of the compounds are respectively
it is difficult to understand why the apparent volumes of the
additional atoms should be so small. The obvious conclusion is,
that the atoms attached by the two additional valency linkages,
possess volumes which are smaller than those attached by the
three others, unless some other explanation is available.
Against this view, apparently, are the compounds
Cl
Cl—P=0
I
ci
Cl
i,
Cl
Cl—p=s
I
Cl
Digitized by v^ooQle
VOLUMES OF PHOSPHORUS COMPOUNDS.
229
from chemical evidence for the volumes of oxygen and sulphur
are those which we might expect.
0 = 8*3 S = 22*0
It follows that in the above respect, the two bromine and
chlorine atoms are exceptional, if this be the correct view of the
matter. Why this should be so is not evident.
One significant fact is, that although the atoms chlorine and
bromine are different in volume, the contractions are about the
same.
This might lead us to suppose that the explanation is, that
the volume of phosphorus is smaller in these pentavalent com¬
pounds than in the trivalent and other pentavalent compounds.
We have, however, no good ground for supposing that an atom
with its five valencies acting is notably different in volume from
that which is true when it only exercises its trivalent function.
For example,
Nitrogen trivalent.
<N=~N>
Vm 31*6 = 2 x 15*8
Cl—N=0
V m 46*8 2«V a 46*0
N x 15*6
Phosphorus trivalent.
Cl
<lCci
^Cl
Vm 93*3 2«Va 93*3
P 27*0
Similar relations have been made out for sulphur, and it has
been shown that S , S iv , and S vi have precisely the same
volume.
The conclusion is that the volume of pentavalent nitrogen
Nitrogen pentavalent.
O
/•
CjH b —N
\
o
Vm 80*4 8 o*2
N 15*6
O
C 5 H n —O—N
\
O
Vm 152*9 2nV« 153*5
N 15*6 O' 7*4
Phosphorus pentavalent.
OCH 3
/
0=P—OCH,
\
OCH,
Vm 1397 2»V a 138-2
P 27-0 O 8-3
Digitized by v^ooQle
LIQUID CHEMICAL COMPOUNDS .
230
and phosphorus possess similar volumes to the trivalent atoms.
The contractions for PC1 5 and PBr 6 are thus not caused by the
pentavalency of phosphorus.
The diminution in volume is thus due to the two additional
chlorine and bromine atoms.
It may be supposed that the compounds PC1 5 and PB1 5 involve
ring structure, the latent valencies of the two additional atoms
becoming active.
Thus the constitutive formulae of PC1 6 and PBr 5 might be
This would also explain why there should be a contraction
in the above instances, and not in the case of
Cl
Cl—P=0 and
/
Cl
The contractions - 7*3 (mean) are not inconsistent with this
supposition, since O s = - 6 * 5 .
The only remaining explanation apparently is that phosphorus
in its action on the halogens, operates differently in the case of
the additional atoms, than on the three ordinary ones. There
are however no adequate reasons for this, that is from the stand¬
point of molecular volumes.
ci
\
Cl— P=S
/
Cl
Thorpe, in his discussion of the series of compounds already
mentioned, supposed that phosphorus was invariably trivalent,
and that the constitutions were
Cl
<v
t,
We have seen that this is not the case, viz. that phosphorus
is pentavalent. The constitutional formulae are those which
have been given.
Cl
:L
S. Cl
Digitized by v^ooQle
VOLUMES OF PHOSPHORUS COMPOUNDS.
231
Only one or two more compounds remain for investigation.
They are
Trimethyl orthophosphate.
OCHj (CH, 0 ) s P 0
0 =P—OCH a V w 1397
OCHo P 27*0
O" 8*3
3O' 103 0 (3 x 8*3)
S*Va 138*3
V« 1397
o
OCH s
k-oc,
l>CH.
see that
Dimethyl ethyl orthophosphate.
PC 4 H n 0 4
V m i6i*8
Comparing the above two compounds we
its normal value.
CH a — 22*1
C a H„
< P—C,H 5
i a H»
Triethyl phosphine.
PC 6 H 15 d 0*812 b.p. 127*5 C 0*55
P 27*0 Vm 170*8
3C a H 5 144*°
S»V a 171*0
Vm 170*8
The value of C was calculated from the preceding compound.
No data for ring compounds including phosphorus are avail¬
able. In all the compounds studied the atomic volume of
phosphorus is the same.
P 27*0
We have thus found the following atomic values
N 15*6 N (aminic) 12 and over
P 27*0
There yet remain the following triatomic atoms to be con¬
sidered—
Digitized by v^ooQle
232
LIQUID CHEMICAL COMPOUNDS .
Arsenic As
Arsenious chloride As Cl 3
fCl
As 4 Cl
Vm 94*4 (Thorpe).
Cl
As
27*8
3 d
66*3
5»V a
94 ' 1
Vm
94'4
Arsenious fluoride. As F s
V« 53*8 (Thorpe).
As 27*8
3F 26*1 (3x87)
2nV a 53*9
Vm 53*8
<F
As-! F
If
Arsenic possesses a volume of only 27*8 in the above
compounds.
What its normal value is we cannot say, probably about 30*5.
Antimony Sb
Antimony trichloride.
f C1
Sb\ Cl SbCl s V m 1007 (Pierre).
lei
Sb 34*2
Cl 3 66*6
2nVa 100*8
Vm 100*7
Antimony tribromide.
f Br
Sb-c Br SbBr 3
IBr
Sb 34*2
3 Br 82*5 (3 x 27*5)
2»V a 1167
Vm ii6*8
The atoms of this group thus possess the following atomic
volumes.
Atomic Volumes of Group 5.
Nitrogen (N) 15*6) A
Phosphorus (P) 27*0 11*4=3 x 3*8
Arsenic (As)
Antimony (Sb)
27*8
34*2
V
2 = 2
x 3*6
Digitized by LaOOQLe
VOLUMES OF PHOSPHORUS COMPOUNDS.
233
Elements of Group 4.
Carbon C.
The question of the atomic volume of carbon has been
exhaustively treated.
It has always been found that the volume of carbon is
C 14*8
at the boiling-point, or a similar one.
If the additive rule were strictly followed, it would no doubt
be shown that carbon has a smaller volume in such compounds
or groups as
CH,
CH,
ch 3 —i— ch 3 ch 3 —C—
than is usually the case, but we prefer to consider any differences
of this nature as being due to constitutive influences possibly
affecting the whole molecule.
Silicon Si.
Si
Silicon tetrachloride.
Cl
SiCl 4
Vm 120*8
Cl
Si
32*0
Cl
4 C 1
88*8
k Cl
—
5 «Va
120*8
Vm
120*8
Siiicon tetrabromide.
Br
SiBr 4
V m 144*3
Br
Si
32*0
Br
Br
112*0 (4 X 28*o)
Br
2«V«
I 44'°
Vm
r 44"3
Vanadium V.
fCl
f C1
'{§
|ci
Vanadium tetrachloride.
VC 1 4
d 0 1*8584 b.p. 154° C 0*48
Vm 120*8
V 32*0
4 C 1 88*8
2nV a
Vm
120*8
120*8
Vanadium oxychloride.
VOCU
fd
V m
106*3
=v\ Cl
V'
32*0
lei
O"
8*3
4 C 1
66*6
2wVa
106*9
Vm
106*3
Digitized by
Google
*34
LIQUID CHEMICAL COMPOUNDS.
Titanium tetrachloride.
III A 1
Cl
Vm
124*47
Cl
Ti
357
Cl
Cl
4 C 1
88*8
2»V a
124*5
Vm
124*5
GermaniunrGe
Ge
Cl
Cl
Cl
IC1
Germanium tetrachloride.
GeCl 4
b.p. 86*o d 18 1*887 C 0*48
Ge
34*5
4 C 1
88*8
123*3
Vm
123*3
Tin Sn
Tin tetrachloride.
Sn
Cl
Vm
i 3 x*i
Sn
42*3
Cl
4 C 1
88*8
Cl
-
Cl
2nV a
131*1
Vm
Elements of Group 4.
131*1
Vanadium (V) 51.
V w 32.
M.W.
Vm.
Carbon (C)
11*0
14*8
Silicon (Si)
Titanium (Ti)
28*0
48*0
32*0
357
Germanium (Ge)
72*3
34*9
Tin (Sn)
n8*o
42*3
Compound.
BoBr 3
BoCl s
Z»(CH 3 ) a
Ztt(C 2 H 5 ) a
Miscellaneous Elements.
V». less n Va.
102*9 - 3 x 27*5
87*0 - 3 X 22*0
72*0 - 2 X 26*0
Il6*7 - 2 X 48*0
Bo 20*4
21*0
Mean 20*7
Zn 20*0
20*7
Mean 20*4
Note. —It will be noted that the atomic volumes of the elements in Group 4 can
be arranged on a regular curve (see Fig. 9, p. 236). The same applies to the elements
of the other groups. Some are notably out of place. For example As should have an
atomic volume of 33 instead of 27 which it is found to possess , and which is similar
to the volumes of Se and Br. I with a volume of 37 is also out of place t and should
possess one of 32. It thus appears that certain unknown disturbing influences exist
which may cause the atomic volumes to be modified. An attempt has been made by
a study of organo metallic derivatives to solve this difficulty. Whilst in certain
cases the expected values have been found , in others the disturbing influences are
again apparent. As has been found to possess in some compounds its true volume.
It is at present impossible to arrive at any definite conclusion till all such compounds
have been investigated. In spite of these disturbances the periodic relation noted
can be traced from a general study of the elements.
Digitized by
Google
CHAPTER VIII.
SUMMARY OF THE THEORY OF MOLECULAR VOLUMES.
A. The Additive Principle.
This feature, which characterizes most physical properties, is
very prominent in molecular volumes, although not, as Kopp
thought, unqualified. This being so, it is essential that accurate
atomic volumes should be found, at least in so far as this can be
done with a property in which the influence of the homologous
increment is felt even from compound to compound in a series.
The numbers determined by Kopp were, for the most part, in¬
accurate, chiefly because the values for the fundamental atoms
carbon and hydrogen were altogether wrong.
The values which have been found for the various elements
in this treatise have been shown to be true, and it is believed
that they are so, for the reason that by their use many con¬
stitutive effects have been revealed and a fairly consistent theory
built up. As regards the series of values, we have come to
the same conclusion as Thorpe, even though he utilized Kopp’s
system in his investigations, that there is a periodic relation to be
traced between them. This periodic relation has been more
clearly shown by means of the new theory. The characteristic
volumes of most of the metals cannot be ascertained at present,
but those of the greater number of the non-metals have been
calculated. The following table shows the periodic relation
indicated:—
The Periodic Relations Between the Atomic Volumes and Atomic
Weights Respectively.
Groups.
i. 2. 3. 4. 5. 6.
H — — — — —
MW. 1 — — — — —
V« 37 - - - -
235
Digitized by v^ooQle
236
LIQUID CHEMICAL COMPOUNDS .
Li.
Be.
Bo.
C.
N.
0 .
F.
7
9
11
12
14
16
19
—
—
20*5
14*8
12*0 \
7 * 4 \
8*7
15*6/
12*0 J
Na.
Mg.
Al.
Si.
P.
S.
Cl.
23
24*3
27
28*3
3 i
32
35*5
32*0
27
2I*6\
25*6/
22*1
K.
Ca.
Sc.
Ti.
V.
Cr.
Mn.
39*0
40*0
44
48*0
5 i
52*5
54 * 8
—
—
—
35 7
32*0
27*4
—
Cu.
Zn.
Ga.
Ge.
As.
Se.
Br.
63*2
65
70
72
75
79
80
—
20
*~
35 *i
27
27
27-8*5
Ag.
Cd.
Sn.
Sb.
Tfc
I.
1077
1117
113*6
n8*8
ng*6
125
126*5
—
—
42*0
[ 34 * 3 ]
37 *o
Au.
Hg.
Tl.
Pb.
Bi.
_
_
1967
200
203*7
206*4
207*3
—
—
—
19 *o
—
46-5 \
48*0
50*1
In the following table and diagram we include all the in¬
formation regarding the atomic volumes of the elements and
their relations with each other, which has been obtained by
means of a study of the molecular volumes of chemical com¬
pounds.
Digitized by v^ooQle
SUMMARY , OF THEORY OF MOLECULAR VOLUMES. 237
We see
(a) That there is a periodic retation between the atomic volumes
of the elements.
(b) There is a tendency for the atomic volumes to diminish in
each series as the atoms increase in weight. The smallest occur at
group seven.
(c) There is a general increase in the atomic volumes of the
members of each group from series one onwards , that is , in the
direction of increasing atomic weight. This increase is usually 3*6
or some multiple thereof.
Variations in the Volumes of the Atoms.
This important subject has already been dealt with but we
here give a summary of the results arrived at. In considering
the variations in the atomic volumes we must distinguish between
the general causes of variation involved when the homologous
increment CH 2 is added, and those special influences which affect
the volumes of individual atoms or, at least, groups of atoms. It
is difficult to separate them.
Carbon and Hydrogen .—These atoms do not vary in volume
except by reason of increase in complexity when the homologous
increment CH 2 is added. For this reason they vary in each
homologous series, and also if we compare the volumes of
hydrogen in corresponding members of different series we also
find differences
H
3-68i
3*671
3*647
These differences may be considered to be due respectively to
the influence of the typical atoms or groups in each series
—H, —I, —COO— and so on.
The latter are influenced by the complexity of the compounds
as well as those of the hydrocarbon chain generally.
In practice it is unnecessary to consider small variations of this
character, except in special cases. Thus carbon and hydrogen
give the average values of
C = 14*8 H = 370.
and these values may be used for all ordinary compounds.
Digitized by v^ooQle
238 LIQUID CHEMICAL COMPOUNDS.
Oxygen O.
The element is difficult to deal with since it varies to a con¬
siderable extent owing to special constitutive influences, and
shows so many different values.
The influence of H and CH 8 is such, that, when in combina¬
tion with oxygen, the volume of O is reduced to perhaps its
smallest value.
The various volumes of oxygen.
O'
O"
>0
"in the OH group (alcohols and phenols)
„ „ (acids)
in the ethers—
" —OCH. group
—OR
in the esters
( doubly bound oxygen O"—
in the aldehydes and ketones
in the acids
in combination with P, S, and N
ring oxygen
(as in ethereal compounds).
Alkyl.
Aromatic.
6-4
r 6
7*4
9*1
7-6
9 * 5 -ii*o
7*4
IO-II
7*4
12*0
8*3- 9*2
7-4-11*0
Sulphur S.
There are two values for this element
(a) S 25*6 (b) 21*6-22*5.
The first is found in the mercaptans and the thio-ethers, also
in those which show differences among themselves in the valency
values of sulphur: —S— as in Cl—S—Cl, in the mercaptans
R—S—H and thio-ethers R—S—R', also in a few inorganic
1
compounds with =S= and =S=.
1
( 6 ) The second value is found in the following groups:
—S—C=N; —N=C=S, also in S=C=S, and ring sulphur.
No other value for =S is found.
Nitrogen N.
There are two characteristic types of nitrogen :—
(a) 15*6 for N=N, and in a few simpler compounds like
NH S , NClg, N a O, N 2 0 4 , and in the groups
—S—C=N, —N : C>, — S-C=N, —N=C=S,
also possibly in others.
Digitized by v^ooQle
SUMMARY OF THEORY OF MOLECULAR VOLUMES . 239
( b ) Amino nitrogen
in the primary amines with the group—NH a , 10*5
in the secondary amines—NHR— and —NR a , 12*0
It has been hitherto supposed that only polyvalent elements
vary in volume, but we shall see that probably others also do so.
Chlorine Cl.
Terminal chlorine Cl' 21*6
Medial chlorine Cl" 24*6
R—Cl' R—CHC 1 "—R'
A number of intermediate values are found.
When attached to the nucleus of benzene two distinct values
for chlorine are known, 19*0 and 21*6.
B. Constitutive Influences.
The physical property of molecular volumes is marked to a
considerable extent by the influence of constitution. The chief
influences which have been worked out in this volume will be
briefly described here.
(a) The Influence of the Homologous Increment
An exact study of the mode of variation of the volumes of
the atoms in a number of series may be made by utilizing the
following principle:—
The relative volumes of the atoms remain the same in the
different members of a homologous series , although their actual
volumes differ from one another .
This has enabled us by means of the relation
v m = ws
where W is the number of hydrogen equivalents, to find S the
volume of hydrogen or its equivalent.
The series of values so found for any homologous series, may
be represented by curves which show the variation of the atomic
volumes throughout the series.
By this means, we have been able to show that the volumes
of the atoms of the first two or three compounds in each series,
and, therefore, those of the compounds, are relatively large. The
values decrease to a minimum at about the fifth member in each
series and then increase in the remaining compounds.
These curves cannot be exactly represented by any formula,
Digitized by v^ooQle
240
LIQUID CHEMICAL COMPOUNDS.
but utilizing the fact that the mode of variation after the fifth
member is nearly rectilinear, we may utilize the following formula
to calculate the volumes of the compounds in a series :—
v#n = w {S + (W - N)K}
W being the number of hydrogen equivalents of a compound.
S the volume of hydrogen which is least in a series.
N the number of hydrogen equivalents in the corresponding
compound.
K the increment in the volume of hydrogen or its equivalent
when hydrogen or its equivalent is added to a compound.
For particular series the formulae are :—
For the normal paraffin series C w H 2n+2 ,
Vm = (6 n + 2) {3*681 + (6 n - 36) x 0*0045}.
For the formic ester series H . COOR,
Vm = (6 n + 7) {3*647 + (6 n - 28) x 0*00375}.
For the alkyl iodide series C*H 2M+1 I,
v m = (6 n + 11) {3-671 + (6« - 24) x 0*00433}.
n is the number of carbon atoms in a compound.
C = 4 H, O' = 2H, O" = 3H, Cl = 6H, Br = 7*6H, I = 10H.
The method may be useful to calculate the volume of a com¬
pound high in any particular series.
Thus, if we require to find the molecular volume of say
heptyl iodide C w H 2n+1 I, we find by use of formula 3 shown above
the following result.
Vm (calc.) 198*7 Vm (obs.) 198*6
The method may be useful in cases which have not been
investigated experimentally—
as e.g. n decane
Vm (calc.) 234*8 V m (by form.) 234*9
using formula 1 above.
In the aromatic series of the phenolic, cresylic, and other
ethers, the benzoates and other series, it is found that the
addition of the homologous increment CH 2 beyond the third,
attached to the nucleus involves a total increase in volume of 37
per CH 2 increase.
Digitized by v^ooQle
SUMMARY OF THEORY OF MOLECULAR VOLUMES. 241
p Cresylic heptyl oxide.
C 6 H 4 (CH 8 )OC 7 H 15
v w 298*3
c (&;
89*6
25*5
0
5*6
c 7 h 15
159*1
279-8
(Cs-C 8 )
i8-S (5 x 37 )
2nVa
298-3
298-3
Unfortunately very few series have been examined with that
degree of accuracy which would make the method generally
useful, but sufficient evidence has been given to show us that we
can extend the scope of our inquiries to fairly complex com¬
pounds, if we possess the requisite material for calculation.
(b) The Influence of Unsaturation .
(i) It is a remarkable fact that, for the most part, unsaturation
produces no apparent effect on the molecular volumes of com¬
pounds, other than that which is occasioned by the subtraction
of atoms. It is true that saturated and unsaturated compounds
differ in volume by about that amount which is due to their
difference in composition. This difference in composition, as a
rule, is two or four atoms of hydrogen.
Unsaturated Compounds.
Compounds.
Vm.
A Vw,
b.p.
C.H 1S
117*8
7.7
38*0
C b H 10 |=|
c»h 8 |=u
110*1
103*9
2 x 7*0 (app.)
39 *o
C.H, . CH 2 . CH,
139*3
7*9
136*5
C e H„ . CH : CH,
131*4
2 x 6-55
146*2
C„H S . C ; CH
126*2
141*6
C s H,OH
81*4
7*3
97*1
CjHjOH
74*1
94*4
C s H 7 C 1
91*9
6*9
46*0
c 8 h 8 ci
85*0
46*0
N(C 8 H,) 8
222*5
3 x 7*2
—
N(C 8 H 6 ) 8
2007
16
—
Digitized by v^ooQle
242
LIQUID CHEMICAL COMPOUNDS.
The average volume of hydrogen in both saturated and un¬
saturated compounds is about 37 and H 2 = 7*4. This is seen
from the foregoing table:—
Comparatively few unsaturated compounds are available for
study, but there seems to be some disturbance, when an un-
saturated atom or group is in association with a | = | or |=|
link.
We have just studied the group
—CH 2 =CH 2 —
(ii) When, however, the group is
—ch 3 =chx—
where X is such unsaturated atom or group there is generally an
increase in volume. For example,
Ethylene bromide CH 2 Br . CH a Br
V m 977
Acetylene dibromide CHBr : CHBr
Vw 92*3
A for H a 5*4 Now H 2 equals 7*3.
The relative increase is therefore
+ 1*9
This can be otherwise calculated
C 2 H 4 Br 2
c 2 h 4
43'2 (2 X 21’6)
c 2 h 2
2Br
54'4 (2 x 27*2)
2Br
2nVa
976
2»V a
V m
977
Vm
36*0
54*4
90*4
92*3
+ i*9
Acetylene tetrachloride.
CHClj . CHClj
b.p. 147*0 do 1*614 C 0*45
Vm 120*5
or CC 1 8 . CHC 1 2 138*2 (Thorpe)
less Cl - 21*6
plus H
5nV a
V m
Ii6*6
+ 37
120*3
120*5
Digitized by
Google
SUMMARY OF THEORY OF MOLECULAR VOLUMES . 243
Tetrachl or ethylene.
CCl, : CCI,
V w 114*8
A for H a 57 H a 7*3
Increase + 1*3
The augmentation is not due to the chlorine atoms directly,
since it is not affected by their number. We must consequently
ascribe it to the ethenoid linkage | = |
Tribrom methane. Tribrom methylene.
CH 2 Br . CHBr a A CHBr : CBr a
Vm 119*3 57 Vm 113*6
The above cases involve a halogen atom on both sides of the
ethenoid link | = |, and in such circumstances the apparent in¬
fluence of these halogen atoms thereon seems indubitable.
Cases occur in which a halogen atom only is found, and the
evidence is here somewhat contradictory.
Ethylidene chloride. Unsymm. dichlorethylene.
CH, . CHCL, CH a = CC4
Vm 88*9 (by obs.) Vm 79*9 (by obs.)
C = 0*46 89 o (by form.) 8o*i (by form.)
* A 8-g
7*2 for H a and A - 1*7.
In this case the volume of CH 2 = CC 1 2 is diminished instead of
increased by the amount 1 7, but there is apparently some dis¬
turbing influence.
It will have been noticed that in the compound CC 1 2 : CC 1 2 ,
the value of A is + 1*3, which thus shows a relative expansion
as compared with the saturated compound.
In the case of the bromine compound, the result is different—
CH, . CHBr a
CH a : CBr a
99*8
6-8
A + i *4
93 *o
CH, . CH a Br
CH a : CHBr
77 * 1
5*8
A + 1*4
7 i *3
In calculating the values for derivatives of ethane, it is neces¬
sary to use considerably larger atomic values, than for propyl
derivatives. This complicates matters. The reason is, that as
we pass from methyl (CH 3 ), ethyl (C 2 H 6 ), and propyl (C 3 H 7 )
derivatives, the change in volume for the addition of CH 2 is
usually considerable. It follows that the volume of hydrogen
will largely depend upon what atom or group is the substituent
of ethane. The volume of hydrogen for the bromine compounds
under consideration is about 3*8.
16*
Digitized by v^ooQle
244 LIQUID CHEMICAL COMPOUNDS.
(iii) It is probable that there is also some volume disturbance
in the groups
—CH(CH 8 ) : CHj— , —CH(CHj): CH(CH S )—
due to the presence of the methyl group.
The effect of other groups, if any, is not known at present.
(iv) It has been shown that in the aromatic halogen com¬
pounds the volumes are similar to those in paraffinoid derivatives.
C e H 5 Cl V w 114*6 Cl 21*8 Cl (paraff.) 21*6
C e H fi Br Vnt 120*0 Br 27*2 Br „ 27*0
C 6 H 5 I V m 130*7 I 37*9 I „ 37*0
Most of the other unsaturated atoms and groups show a
diminution when attached to the radical C 6 H 6 . Halogen com¬
pounds apparently do not. If however we suppose that in the
group
^C—Cl
JL
the usual diminution of - 15, due to the combination of
chlorine with the C 6 H 6 group, be neutralized by the augmenta¬
tion + I'S, caused by association of chlorine with the | = |, we
have a possible explanation of the exceptions to the rule just
noticed.
We notice that the apparent volume of iodine is considerably
larger in C„H 5 I (as in allyl and propargyl iodides) than in
saturated derivatives.
The effect of the increasing magnitude of the substituent is
seen in the compounds just mentioned.
Propyl compounds.
Vm.
A.
V m .
Allyl compounds.
CH g . CH a . CH a Cl
91*4
7*2
84*2
CH a : CH . CH 2 C 1
CH 8 . CH a . CH a Br
97*4
6*9
90*5
CH a : CH . CH a Br
CH 8 . CH a . CH a I
106*9
57
101*2
CH a : CH . CH a I
CH 8 . CH a . COOCH 8
104*2
5*8
984
CH a : CH . COOCHj
CH 8 . CH a . CH a Cl
92*0
777
Propargyl compounds.
CH : C . CHjCl
CHj . CH a . CH a I
107*1
12*5
94*6
CH : C . CH a I
A thorough investigation of all these phenomena would
enable us to solve many problems of constitution, such as, for
instance, differentiation between compounds of the types
CH S . CH : CH a X and CH a : CH . CH a X
Digitized by
Google
SUMMARY OF THEORY OF MOLECULAR VOLUMES . 245
(c) Partial Ring Structure .
Intermediate between open and closed chain compounds are
what are known as
Partial or Incomplete Rings.
In order that they may be formed, it is necessary that two
atoms or groups, which may be supposed to possess residual
affinity, occur in a hydrocarbon chain attached to different carbon
atoms. If the carbons are near to each other, there is no need to
suppose any variation from structure usually considered when
the tetrahedral arrangement of the valency links of carbon is
understood. The additional feature of curvature of the hydro¬
carbon chain may also be the normal condition of things. In
the case of saturated compounds the plane formulae are figured
thus —
CH 2
CHj
1
J
CH 2 -X N
I )
CH- Y /
ch 3
If the two attracting groups are united to carbons not in the
immediate neighbourhood of each other, we must suppose that
they are brought near by the curvature of the hydrocarbon chain,
or the alternative supposition just given is true. In any case such
a structure affects the molecular volumes of compounds.
Thus—
CH^
-X\
CH„-
)
CH.
-T
CH-
~X\
\
I
J
If the compound be unsaturated, it is necessary to suppose
two forms which may thus be represented—
CH-
X-CH
II
CH-
_ Y /
II
CH-Y
The syn compound.
The anti-compound.
The former would be smaller in volume.
Digitized by v^ooQle
246
LIQUID CHEMICAL COMPOUNDS .
We must also suppose that in the benzene di-substituted
molecule the ortho form would be
\
This, however, according to the tetrahedral arrangement of the
carbon links, could be given two types of arrangement: one in
which the atoms or groups appeared on the same side of the
ring and one in which they would be disposed on the opposite
sides.
It should be remarked that in the saturated chain there is no
hindrance to the two reactive groups occupying the sign position,
so that only one form is possible. In the unsaturated chain the
possible rotation about an axis joining two carbon atoms is pre¬
vented by the residual affinity associated with an ethenoid link.
Two forms are thus possible.
The attracting atoms or groups are the halogen atoms
Cl, Br, I, the hydroxyl group OH, the amino NH 2 , etc., the
nitro N 0 2 , the carboxyl COOH, and perhaps other groups.
The contraction
for a /9 compounds is about - 3*0
1 > ay »» >f “4*5
In ring-compounds the ortho arrangement has been specially
studied (q.v.).
(d) Ring Structure.
This subject has already been dealt with at some length, but
the chief points can here be stated.
Ring compounds are invariably characterized by contractions.
This is due to a diminution in the individual atomic volumes
forming the nucleus, and also in the case of the associated
hydrogen atoms, when such are present.
The compounds ( a ) benzene and (b) hexamethylene have been
studied not only at the critical and boiling-points but also at
corresponding pressures, and it has been shown that the relative
volumes of the atoms remain the same as in straight-chain
compounds .
Digitized by v^ooQle
SUMMARY OF THEORY OF MOLECULAR VOLUMES. 247
The contractions for single-ringed compounds are
□a - 6*5 D4 - 8 *5 Ds - IX *5 □« - !5’o □? - 20-0 Os - 25*5
Double and triple rings have contractions
I I I 6+ 5 “ 26*0 I | l a x 6 - 30*0 camphor, etc. - 31*0
1 I 1 la x 6 - 4 8 *° IM 1 * x s + b “ 45 *°
Not many independent double-ring compounds are available for
study.
(e) Molecular Volume and Valency .
It has been supposed that there is a connection between
volume and valency. However this may be in a general sense>
there is no clear connection between the volume and the funda¬
mental valency. When, however, we compare the atomic
volumes in the different series of the periodic arrangement of the
elements, we find a general indication of a similar increase from
series to series. For example:—
Averages.
Series I
C 14*8
N 15-6
0 ii’o
F 87
A
+ 17*2
+ 14*6
+ n *4
+ 13*4
I 4 *I = 4X37
Series II
Si 32*0
P 27*0
S 22-0 \
25*6 J
Cl 22*0
M. 23*8
A
+ 37
+ 5*o
+ 3*6
+ 4*9
4*3 = i x 4*3
Series III
Ti 357
V 32*0
Cr 27-4
Br 27*0
Ge 35-1
As 27*0
Br 27*0
A
+ 6-g
+ 7*3
+ 10*0
8 # o = 2 x 4*0
Sn 42*0
34*3
—
I 37-0
These differences vary somewhat, but they are similar. It is,
however, probable that the numbers themselves, which vary con¬
siderably under different circumstances and with various environ¬
ments, are not quite the significant volumes. In order to arrive
at a generalization of more than qualitative significance, it is
necessary to know precisely what the term molecular volume
stands for and it is also necessary that the atoms should be
placed under comparable conditions. How much of the real
molecular volume space consists of atoms with their shells, and
how much is made up of their movements of vibration we do not
know. It is possible that some simple relation between volume
and valency, similar to that discovered by Barlow and Pope, might
be discovered if we could divide up the apparent volumes into
their real parts. This may one day be possible, and the relation,
one of simple proportionality, discovered.
Digitized by v^ooQle
248
LIQUID CHEMICAL COMPOUNDS .
The unit 37 seems to be significant for (a) several atoms are
integral multiples of this unit—
C 14-8 = 4S h = 37 = s o = n*o = 3S s = 25-6 = 7S.
Also (b) several atoms of varying atomic volume seem to
show volumes which differ by just this unit—
7*4
S 22*0
N 12*0
11*0
25*6
15*6
3*6
3*6
~6
The possibility of variation in the atomic volumes cor¬
responding to variations in acting valency thus occurs.
It has been thought that as the acting valency of an atom
changes its volume varies also. This is found not to be the case.
Thorpe has stated this in the course of his work on molecular
volumes. Thus: “The inquiry affords us no evidence in favour
of the hypothesis that the specific volume of an element in com¬
bination is modified by any possible variation in the affinity value
which it may possess. The observation of compounds of sulphur
and phosphorus appear conclusive on this point.”
We do not know how Thorpe came to this conclusion, since
his formulae make phosphorus trivalent and sulphur divalent—
C \
P—0C1 Cl—S—0C1 CIO—S—0C1
• Cl^
With the accepted formulae, which are confirmed by means of
our theory, we find that phosphorus and sulphur may act with
varying valencies. There is, however, no change in the atomic
volumes of the elements due to this cause, as Thorpe supposed.
Cl
/
Cl—P
\
Cl
or P*“ 27*0
It should be mentioned that Prideaux* values for PC 1 5 and
PBr 6 seem to lead to a different conclusion, but his observations
are capable of another explanation (q.v.), and in any case are
exceptional.
Cl
/
0=P—Br
\
Cl
P v 27*0
Digitized by
Google
SUMMARY OF THEORY OF MOLECULAR VOLUMES . 249
Cl—S—Cl
S" 25*6
Q=N—Cl
N'" 15*6
Cl
/
o—s
\
Cl
Siv 25*6
C.H^O—N
O Cl
\ /
s
^ \
O Cl
Svi 25*6
o
\
Nv 15*6
We do not possess the requisite data for any other elements, but
the results just obtained appear conclusive on this matter.
It is true that the volumes of the elements vary, but this
variation cannot be shown to be due to their variable valency
relations, but to some other cause.
S" 21*5 and 25*6
N"' 12 15*6
O" 7*4 n*o
Br 27*0 to 28*5
(/) The Constitutive Influences Due to Groups.
The question of special influences due to groups is a very
important one, and for thorough treatment many more accurate
data are needed than are at present available. In so far as the
homologous open-chain series are concerned, the additive rule
may, for practical purposes, be said to prevail, unless the com¬
plexity be greater than a certain amount.
Volumes of Groups in Organic Compounds .
CH S 26*0 C 2 H a 48*0 C 3 H 7 70*0 C 4 H 9 92*3 C 6 H a 92*8
OH io*o CHO 25‘9 CO 22*0 COOH 37*8
NH 2 17*4 N 0 2 32*0 CN 30*4 S. CN 52*0
The volumes are average ones for the groups in question, but
they do not vary very much under ordinary circumstances except
perhaps for the group CH 3 . Taking the average volumes of the
atoms already indicated, we find how the volumes of the groups
CH 3 and C 2 H 5 may vary according to the particular typical atom
or group to which it is attached. The ethyl CH 3 group varies
most, the C 2 H 5 group to a lesser extent, and both vary most
Digitized by v^ooQle
250
LIQUID CHEMICAL COMPOUNDS .
when H,OH and Cl are the other groups making up the com¬
pound. These variations are due to causes not quite understood
as yet, but probably complexity is one of them.
It is also possible that there may be some interactions be¬
tween the alkyl and other groups. An important example of
variation in the values of the groups is found when they are
attached to aromatic radicals like phenyl C 6 H 5 . The volume
of this group is taken as 92*8, and the volumes of the other
groups as in the preceding table.
The Contractions Found when Unsaturated Groups are in
Union with the Phenyl Group C 6 H 5 .
Group.
A.
Group.
A.
Group. A.
—OH
-0*9
—COOH
- 3*9
CH S and \
—CN
-1-6
—CH 2 C 1
-i *9
other alkyl > - 0*5
-nh 2
“ 4*3
—no 2
-27
groups *
—COC 1
-i-8
—PC 1 2
- 2*0
These variations may be ascribed to some interaction between
the phenyl C 6 H 5 radical and the unsaturated groups. For this
reason the formulae, representing this constitutive action, may be
shown generally as—
^ CH \
H C—-X
CH CH
ch/
X representing the unsaturated group.
It is easy to show that unsaturation is the cause, because
when the hydrogen in the OH and COOH groups are replaced
by alkyl groups like CH 3 the contractions disappear.
Phenol
Anisol
C,H 5 . OH
C,H S . OCH,
Vm
102*0
125*4
2 «Va
102*9
125*2
A
-0*9
Benzyl alcohol
Anisol
C 4 H„ . CH,OH
CjH e . OCH,
1237
125*4
125*2
125*2
-i *5
Benzoic acid
Ethyl benzoate
C 6 H 6 . COOH
C,H, . COOC,H,
126*9
x 74*4
130*6
174*1
- 3*7
Phenyl propionic acid
Ethyl benzoate
C„H,. CH, . CH,COOH
CjH,. COOC,H,
170*9
174*4
174*1
174*1
- 3*2
Digitized by
Google
SUMMARY OF THEORY OF MOLECULAR VOLUMES . 251
It has already been shown that the group NH 2 is responsible
for a contraction of - 4*3. It is remarkable that the groups
—N(CH 3 ) 2 , —N(C 2 H 5 ) 2 produce the same contraction, a fact
which points to the common atom nitrogen as being the part
affected. If the group be unsaturated, e.g. N(C 8 H 5 ) 2 , the contrac¬
tion is apparently greater.
Not many other groups have been sufficiently studied, but
cases which are subject to such contractions may be given. For
instance, if we compare
R—OH and R—O—R
R—NH a and R-NH-R or R^=N—R
we find that
O = 6*4 and io*o
N = 10*5 and 12*0
Such cases are, however, difficult to adequately account for. This
is one of the great difficulties which attend the study of the mole¬
cular volumes of open-chain organic compounds, for it is not
quite clear whether and to what extent the individual groups
making up these compounds influence one another. Such a
study is easy in ring compounds such as benzene derivatives, and
under such circumstances a fairly complete and reliable analysis
has been made.
C. A discussion of the special conception embodied in mol¬
ecular volumes, and the relation which obtains between
this property and other physical properties such as boil¬
ing point, surface tension and viscosity.
A discussion of the special conception embodied in molecular
volumes, and the relation which obtains between this property
and other physical properties such as boiling point, surface ten¬
sion and viscosity.
It is hardly possible in a formal treatise like the present to
altogether avoid dealing to some extent with the physical mean¬
ing of molecular volumes, and the derived conception of atomic
volume.
Digitized by v^ooQle
252
LIQUID CHEMICAL COMPOUNDS .
If the specific volume = - in c.c., then the molecular volume of
a
say pentane C 5 H 12 is
g- MW 72
0'6l2 ——=—- = 117-6
d o # 6i2
at the normal boiling-point
The molecular volume in the first instance is the volume of
that weight which represents the weight of the molecule, or the
molecular weight multiplied by the specific volume. This volume
under well-defined conditions such as those obtaining at the boil¬
ing-point may be supposed to bear to the real molecular volumes,
a relation, which is the same for all substances, and thus pro¬
portional to the real molecular volumes. Jhis was assumed by
Kopp when he started on his investigations of the molecular
volumes of liquid organic compounds at the boiling-point, and
this has been justified by subsequent work.
It has been shown that
Vb.p. = 3 / 8 V.
but Vo = V 4 Vc
... v b .p. = * x a Vo = s / 2 Vo
that is, the volumes of the most various liquid compounds are,
at the normal boiling-point, one and a halftimes the volume at
absolute zero. This volume V 0 is consequently proportional to
the real molecular volume, and so are also the volumes at the
boiling-point V b . p ..
If the law were strictly followed, the investigation of the
molecular volumes of compounds would be a simpler matter
than it is. It depends upon the fact that the internal pressures
are the same for all compounds at the boiling-point, subject to
small variations due to differences in complexity and other
causes. These disturbances destroy the simplicity of the rules
regulating the relation between the atomic volumes to the whole
molecular volume. Young, who investigated a number of organic
substances minutely, from the point of view of Van der Waal’s
equation of condition, attributes the departures from Van der
Waal’s theory to the constitutive influences.
We have reason to believe that in liquids the molecules exist
under an intense intermolecular pressure, which affects the boil¬
ing-points of the compounds, their surface tensions, and their
viscosities. These intermolecular affinity forces act in such a way
as to oppose the heat forces, which consist of vibratory movements
Digitized by
Google
SUMMARY OF THEORY OF MOLECULAR VOLUMES . 253
of some kind. The intermolecular forces, presumably of great
magnitude, together with the heat forces serve to limit the
molecular motions and define the space called the molecular
volume. If we consider simply those portions of the molecules
called the nuclei, measured by 3V D , in the Lorenz-Lorentz formula,
or considered in the Clausius theory, it is impossible that they
should completely fill the space occupied known as the molecular
volume. There must of necessity be a molecular interspace as
well as possibly atomic interspaces. This interspace is measured
by V - b according to Van der Waal’s formula.
We may now consider the question :—
(a) Of what constitutes the atom and consequently the
molecule, and
(< b ) What is the nature of the molecular movement ?
If the Boscovitchian hard atom cannot be entertained, owing
to the necessity of accounting for the elastic properties of matter,
we must suppose that the nucleus just referred to is enclosed
within a shell of dielectric. We consider in the first place that
the molecule is made up of the atomic nuclei and its real volume
is equal to their sum. The atomic nucleus is however not the
whole of the atom, for if we were to cool a substance to absolute
zero we should find that it occupied a volume considerably in
excess of that of this molecular skeleton so to speak. In other
words the term b of Van der Waal’s formula is considerably
larger than 37 b of the Clausius theory. Traube supposes that,
b = 3-5 or 4 2V D .
He iemploys the molecular refraction as a measure of the
nuclear volumes of the atoms contained in a substance, and
moreover calculates that the molecular refraction of a saturated
compound is proportional to the number of valencies of the com¬
ponent atoms. At any rate this atomic shell represents that
portion of the atom which is permeable to light, and constitutes
that dielectric medium which enables electro-magnetic radiations
generally to traverse the liquid at a speed which is characteristic
of the particular liquid under consideration. The molecular volume
at absolute zero is thus equal to the sum of the atomic nuclei plus
the sum of the dielectric shells . It may be supposed that the
chemical forces which bind atom to atom to form a molecular
combination, is accompanied by stresses and strains in this
dielectric.
Digitized by v^ooQle
254
LIQUID CHEMICAL COMPOUNDS.
We are now asked to consider a space equal to one and a half
times the volumes of all the atomic nuclei in the molecule:—
since Vb.p. = $ Vo = $ b.
That this space is occasioned by motion of some kind, is obvious,
but on the nature of this motion there has been much dispute.
One consideration, which may prove of considerable importance,
has never been commented on to the best of our knowledge.
It is this,—why may not the dielectric medium itself expand
with rising temperature and fill up the extra space occupied by
the liquid caused by the separation of the molecular and also the
atomic centres ? It is not difficult to imagine such a condition.
We may suppose for instance that the dielectric consists of a
series of shells of diminishing magnitude as we approach the
centre, and related in such a way that a central section would
show a spiral originating from the atomic nucleus. Energy could
become latent by the “ solid spiral ” contracting as a watch spring
contracts when it is wound up, and moreover could vibrate by a
repeated winding and unwinding within limits dependent on the
stress within the medium as an ordinary spring might do. Such
a motion of the shell, combined with the resultant motion of the
nucleus, would constitute the heat motion. We do not insist on
this particular mechanism, but give it for what it is worth. The
idea of a dielectric expands as the temperature rises, is, however,
worthy of every consideration, for we know that the index of
refraction fi , and therefore the velocity of light in the particular
liquid medium, is a function of the temperature. It may be stated
that the dielectric shells might not be susceptible to differences
in density from compound to compound or for one compound
under various conditions, in the same sense that matter in general
differs in density. The differences might consist of different
degrees or kinds of strain.
If we consider the whole molecule and its expansion under the
influence of temperature, we see that there is every reason to sup¬
pose that under circumstances of equal internal pressure or under
corresponding conditions, the volumes of the molecules should be
proportional to their absolute zero volumes and thus of their real
molecular volumes. It is however possible, and indeed probable,
that as we approach the critical point, certain disturbances might
occur which cause departures*from ideal conditions, and from the
more simple conditions which occur at or near absolute zero.
Digitized by v^ooQle
SUMMARY OF THEORY OF MOLECULAR VOLUMES . 255
This, in effect, is the result of the remarkably short range of the
affinity forces, and the consequent rapid variation with the distance
between the molecular centres. It follows that constitutive
features which are apparent at or near the melting point are to
some extent marked at and above the boiling point.
The Visibility of the Liquid.
In our view, one circumstance of great significance is, that,
when a vapour condenses it becomes visible. Steam, for example,
condenses to a transparent but visible liquid water, which, except
for conditions of mobility, maintains certain characteristics be¬
low the melting-point and in the solid state which are noticed
in the liquid. The clear limpid water solidifies to the clear
glassy ice and this characteristic is maintained down to absolute
zero. For that matter an opaque liquid like mercury maintains
this characteristic below the melting-point, and so far as we know
even to absolute zero. In both cases, however, the vapours are
transparent. The reason is, that in the vapours, the light passes
through the interspaces which separate the molecules, but in liquids
when transparent, it passes through a different medium, the dielec¬
tric shells of the atoms and possibly of the molecules as a whole,
and this condition is maintained down to absolute zero. We
believe that the features just noted can only be explained on the
assumption of compact structure in the sense indicated. It is also
to be remarked that the theory of absorption spectra implies that
light traverses a medium different to that of the external ether.
It is not difficult to understand why Richards should have
come to the conclusion that the atoms in a liquid as in a solid
molecule are in actual contact with each other. We think that,
it is necessary to go a step farther and to suppose that the mole¬
cules as just defined, are also in actual contact. It is thus
found that liquids are compact aggregations of molecules of de¬
finite and characteristic compressibility under well-defined condi¬
tions, and that the compressibility of the liquid as a whole, measures
the compressibility of the dielectric associated with the atomic
nuclei. The question of the mobility of the molecules under such
circumstances will be referred to immediately.
Another idea of considerable moment, and one derived from
a study of molecular volumes, is that if there is considerable
motion of translation of the molecules with its consequent impacts,
then these motions of translation will depend upon the molecular
Digitized by v^ooQle
256
LIQUID CHEMICAL COMPOUNDS.
weights. We should thus expect that a heavy molecule like
Ci 6 H 84 , would slow down as compared with a light one like
butane (C 4 H 10 ) or pentane (C 5 H 12 ).
According to the kinetic theory
pressure p = MNV 2
M representing molecular weight, N the number of molecules, V
the velocity.
If we compare two molecules of different complexities then
for 1 gr. mol. of each substance
P\ _ Mi Yh
/ a “M a vv
Since at the boiling-point p x = p 2
Mi_YV r
M a ~ V\ or
(velocity)* 00 moIecul ^ weight -
Thus a small molecule like C 6 H 10 might be expected to
occupy a large volume as compared with a heavy one like
QoH 84 .
We find that the opposite is the case. Pentane C 5 H 12 at the
boiling-point occupies a volume of 117*8. A similar complex in
octane C 8 H 18 occupies a volume of
32 x 3 725
= 119*2
The difference would be much greater if hexadecane C 16 H 34
were under consideration.
The Motion of a Molecule through the Liquid Mass .
It is not difficult to imagine that, under such conditions, an
aggregation of elastic substances, endowed with enormous amounts
of kinetic energy, might diffuse or migrate excessively slowly
against the intense fields of force which are consequent on the inter-
molecular residual affinities. We cannot conceive of the existence
of the excessively slow motion characteristic of the molecules, even
in dilute solutions, on any other assumption than the one already
made. The motion of a molecule in a liquid mass may be compared
to that of the motion of a comparatively large mass (like a projected
bullet) through a viscous medium like tar.or treacle, as compared
with its ordinary motion through the air. The analogy is still
Digiti , iy v^ooQle
SUMMARY OR THEORY OR MOLECULAR VOLUMES. 257
incomplete since the bullet has no affinity for the medium through
which it is passing.
It may be remarked in conclusion that a consideration of the
molecular volumes from the present point of view is really inde¬
pendent of these theoretical questions, and depends only on the
empirical rule that the volumes at the boiling-point are for the
most various substances four times the volumes at absolute zero.
It now remains for us to consider how it is that the molecular
volumes reflect so accurately the chemical constitutions of sub¬
stances. In dealing with molecular volumes, it is necessary to
take into account such properties as boiling-point, viscosity, and
surface tension. We should however note that boiling-points do
not always indicate differences in chemical constitution, nor show
results which are evident from a study of molecular volumes. For
instance variation in molecular volume and boiling-point are not
affected to the same degree, by such constitutive features as ring
structure.
Dipropargyl C 6 H 8 .
CH : C . CH 2 . CH 2 . C : CH
Volume m*o A - 15*0
Boiling-point 85*0 A - 47
Benzene.
c 6 h 6
96*0
80*3
The variations of the boiling-points for a given chemical
change are usually greatly in excess of the volume changes, but
not in this case. On the other hand, differences in boiling-point
are often accompanied by well-defined changes in volume, but
the latter are most frequently in the opposite direction—that is,
when the one is positive the other is negative. The intermole-
cular forces are doubtless of considerable importance in deter¬
mining the molecular volume, and these may be considered to be
determined by the forces of residual affinity arising from the in¬
dividual atoms. It has been shown that the range of the mole¬
cular activity or influence is equal to the distance which separates
two molecular centres. The molecular forces influence such
physical properties as
(a) The boiling-point .—For example, that compound which
possesses the highest boiling-point is obviously the most difficult
to separate from its neighbours, and to endow with the kinetic
energy necessary to cause vaporization. As already stated the
effect upon the molecular volume is generally in the opposite
direction to that upon the boiling-point. This is seen in such
cases as
17
Digitized by v^ooQle
253 LIQUID CHEMICAL COMPOUNDS.
CH, . CH a . CH a Cl
b.p. 46-5°
A - io°
(CHj) a CHC1
36*5°
Vm 92*0
A + 2*3
94*3
CHj. CHBr a
b.p. in-o
A + 19*3
CH a Br . CH a Br
130*3
V M 99*o
A “ 2*0
97*o
CH S
p c,h 4 <^
CHj
b.p. 138*0
A + 3*9
CH,
m C,H 4 <^
CH,
141-9
V w 140*2
A - 2*3
l 37'9
The explanation which seems to best suit these cases is that the
intermolecular forces are most intense in the higher boiling-point
liquids (when two or more isomers are compared), and it is evident
that the peculiarity which occasions this difference, is one which
favours a more striking manifestation of intramolecular force
than in the other case. It is natural then to suppose that the
particular combination or spatial association of atoms or groups,
or their distribution throughout the molecule, are the factors
which cause the differences in the various cases. It should,
however, be clearly understood that changes in the boiling-point
are not always evidence of corresponding changes in the (a) mole¬
cular volumes , (b) surface tension , and (c) viscosity.
It is natural to suppose that boiling-point, surface tension,
and viscosity are parallel phenomena, for those influences which
tend to increase or diminish one property, also influence the
others. Thus we find :—
b.p.
Viscosity
Surface tension.
Vf X IO 6 .
N x 1000
n-hexane
68*6
221
-6*6
-13
iso-hexane
62*0
208
xylene 0
141
254
16*0
-2*0
-21
-0*1
xylene m
139
233
15*9
- 1*0
±0
-0*1
xylene p
233
15-8
We see that any explanation of the peculiarities manifested
by the molecular volumes of compounds must take into considera¬
tion the part intermolecular forces play in determining the mole¬
cular volumes of compounds as well as the intramolecular forces
themselves. The latter may possibly in some cases be in the
same direction as the intermolecular forces, or on the other hand be
contrary to them. It probably is not easy to make these distinc-
Digitized by v^ooQle
SUMMARY OF THEORY OF MOLECULAR VOLUMES . 259
tions in practice, nor clearly in every case to analyse the complex
conditions which obtain. Some constitutive influences are evid¬
ent enough, such as those of ring structure, the contiguity of
groups or separation of groups in dihalogen derivatives of the
paraffins, and the ortho, meta, and para modifications of the di¬
substitution products of benzene, etc. The work which has been
done in this volume, whilst to some extent useful for the ex¬
planations given of some of the phenomena, is probably of
greatest value for the analysis of the effects themselves apart
from their causes. The explanations will come later. Even
as regards the effects, data are very meagre in many cases, and
though the numbers obtained by means of the formula de¬
scribed in Appendix II. may be exceedingly useful as preliminary
determinations of data, yet they can never carry the weight of the
results of well-directed and well-executed experiments. It is to
be hoped that the latter will soon largely increase.
There is no doubt that in spite of the care taken, many parts
of the present theory may have to be altered later as data ac¬
cumulate, and as our knowledge of the physical property increases.
The identification and explanation of constitutive effects is not
always easy. Some particular atomic values—generally those
found in the homologous series R—X are taken as standard,
and by the method of summation the value 2 nV a is found. The
difference V w — 2 riV a then measures the constitutive effect.
Sometimes a mean atomic value is taken, and it then follows that
no account is taken of the variations. The great difficulty is to
identify the effect with a particular atom or group. When this
seems possible, it sometimes happens that other atoms or groups
might equally well be identified with the effect in question. Only
a careful examination of a large number of data can overcome
these difficulties. It will generally be found that the constitutive
effectsV w — S n V a are traceable to some modification in particular
atomic values, and a considerable advance is made when we are
able to ascertain for certain which atoms are marked by the
variation in question, and by how much.
For instance, the contraction for ring compounds like benzene
C 6 H 6 amounts to -i5 # o. Careful inquiry shows that this is
made up from a contraction of all the atoms in the nucleus
C« 12*8 instead of 14*8 and H = 3*2. It follows that we know
the volumes of such residues as C 6 H 5 —, C 6 H 4 = , etc., and so are
17*
Digitized by v^ooQle
26 o
LIQUID CHEMICAL COMPOUNDS.
able to find out the volumes occupied by the side chains. For
instance we show that the volume of CH 3 is the same in the
homologues of benzene, if sufficiently separated.
Vw.
Benz. Rud.
«CH a .
CH a
c.h b .ci
Eij 118-25
92-8
—
25*45
«,), 140-51
89-6
50-91
25*45
1:3:5 C,H».jC
Hj) 3 162-75
86-4
76-35
25*45
This knowledge may serve as a basis for the determination
of the volumes of components of other aromatic compounds and
also enable us to measure constitutive effects.
OC 6 H 4 (CH 3 ) a Vw 137*6 C 6 H 4 89*6
2CH3 50*9
2»Va 140*5
Vw 137*6
A For ortho struct . - 2-9
It happens in this case that SnV a is similar to the volume of
p. xylene. In other cases this additional information may be
lacking, and we depend on the regularity already observed.
Greater simplicity is introduced into the work, for our guiding
principle by being able to assume that in all these aromatic com¬
pounds the phenyl or similar radical, undergoes no change of
volume in the different combinations. This enables us to deal
with fairly complicated ring compounds.
Much greater difficulty has been experienced with open-chain
compounds, owing to our lack of ability to decide which atoms
or groups take part in the constitutive effect. This is largely
due to the fact that many constitutive effects are more or less
affected by the general influence of complexity, which is not de¬
pendent on any simple law.
It may also be stated that one reason why the effects may
differ somewhat, in degree at any rate, under other conditions than
those at the boiling-point, from the effects observed at this point,
is just the effect of this temperature factor. We might expect
that constitutive effects should be more prominent say at the
melting-point than at the boiling-point, because in addition to
the diminishing influence of the heat forces, the forces of affinity
have increased according to some high power of the distance
between the molecular centres, and may even change sign at
some point It is also true that the effects may be different
from those observed from the examination of other physical
Digitized by v^ooQle
SUMMARY OF THEORY OF MOLECULAR VOLUMES . 261
properties such as refractive power, and magnetic rotatory power
for just the reason that they do not possess a temperature factor
as do molecular volumes.
In consequence of the facts just mentioned, it is hoped that
any deficiencies or irregularities found, will be leniently dealt
with, and regarded as part of that unavoidable element of crudity
which attaches to all theories in their initial stages.
Digitized by v^ooQle
Digitized by v^ooQle
APPENDIX I.
A FORMULA BY MEANS OF WHICH THE MOLECULAR VOLUME AT
THE BOILING-POINT MAY BE CALCULATED.*
In an extended study of molecular volumes, it was found neces¬
sary to calculate some of the values at the boiling-point, and the follow¬
ing formula has been found suitable for this purpose:—
t-‘~ ('-!£)■
The only data necessary are the density at o° and the boiling-point.
The value of c is given in the following table, the data being those
of Thorpe:—
Table of Values (Inorganic Compounds ).
Compound.
B.p.
do .
dB.p.
c .
V*.
calc.
Vw.
obs.
Per cent
error.
saci 2
• 139-6
1*85846
1*60610
0*460
135*5
135*5
+ 0*0
SO,Cl. OH.
• 155-3
178474
1*54874
0*420
75*5
75*05
+07
SO,Cl a
AsC 1 3 .
. 70*0
170814
1*56025
0*462
86*3
86*3
+ 0*0
. 130*2
2*20500
1*91813
0*447
95 *i
94*37
+ 0*76
AsF 8 .
VOCI3 .
. 60*4
2*6659
2*4497
0*490
53*6
53*84
-0*44
. 127*2
1-86534
1*63073
0*452
106*5
106*2
+0*3
POBrCl a .
• 137-6
2*12065
1*83844
0*458
107*5
107*4
+ 0*1
PSClj.
. 125-1
1*66820
1*45599
0*455
116*3
ii6*i
+0*2
POCI3.
. 107-2
171163
1*50967
0*474
101*0
101*4
-0*4
JiCL . .
• 136-4
1*76041
1*52223
0*460
124*5
124*5
+ 0*0
SiCl 4 .
• 57-6
1*52408
1*40294
0*500
120*2
120*8
-o *5
N 2 0 4 .
. 21*6
1*49030
1*43958
0*473
64*0
63-9
+ 0*0
Mean value, 0*463.
It is found that by means of the above formula the volumes of
compounds of a similar order of complexity can be calculated to within
1 per cent.
273 d °
Example.—GeCl 4 , d w 1*887 (Winkler), b.p. 86*o°, = 0*760, y = 0*46 x 0*240
1*887
= 1*104, <*b.p. = = 1*709. M.W. = 213*8; V w . 125*1.
Observed, C 14-8, Si 32, Ge [36*3], Sn 42*3, Ti [357].
The formula can be used indifferently for inorganic and organic
compounds, but the value of c in the latter varies somewhat as the com¬
pounds vary greatly in complexity and the chains lengthen.
* Adapted from “ Joum. Chem. Soc.”
263
Digitized by v^ooQle
264
LIQUID CHEMICAL COMPOUNDS,
The value of c for organic cyclic compounds without side-chains is
similar to the above.
Cyclic Compounds .
Compounds.
M.p.
oro°.
B.p.
du
o°
or m.p.
d r
T=B.p.
c .
Vm.
calc.
Vm.
obs.
Per cent
error.
C 8 H 6 (Benzene)
C 4 H 4 S (Thiophen) .
C 10 H 8 (N aphthalene)
C 10 H 14 (Hexahydro-
naphthalene)
C 14 H 10 (Phenan¬
6-o°
8o°
0*8940
0*8133
0*470
957
96*0
-0*03
0*0
84
1*0884
0-9874
°*434
85*5
85*0
+ o*6
79-2
217
0*9777
0*8674
0*456
147*3
147*2
+ 0*0
0*0
200
0*9419
07809
0*487
170*0
171*2
-07
threne .
100*5
340
1*0630
0*9073
0*440
197*8
195*2
+ i # 3
C 6 H 5 N (Pyridine) .
C 9 H 7 N (Quinoline) .
0*0
0*0
”5
234
1*0033
1*1081
0*8826
0*9211
0*462
0*439
89*3
141*1
« 9'3
I40*0
+ 0*0
+o*8
Mean value,
0*460.
The results of calculation show very fair agreement with observa¬
tion, and thus giving a fairly trustworthy method for the calculation of
unknown values.
Example.—Hydrindene , C 9 H 10 : d 5 0-957, b.p. 176°, £=46, V w .
144-0. SnV a =*(4x9+ 10)3-7 = 46 x 37 = 170-2, A = - 26-2
for ring. Contraction for 1 six-membered ring + 1 five-membered ring
= - 15 “ n *5 = “ 26 5.
Acenaphthene , C 12 H 8 .—The value of c for phenanthrene is 0*440.
For acenaphthene, d lQZ 1*030, m.p. 103°, b.p. 277 0 . V m 166*6,
2 «V a 207*2, A = - 40-6. Contraction for 2 six-membered rings + 1
five-membered ring = - 30-0 - 11-5= -41*5.
The only difficulty is met with in open-chain organic compounds.
For compounds like chloroform, carbon tetrachloride, and trichloro-
methane, the value of c mentioned above (0-460) may suffice.
* c, in general, increases by 0*024 for every addition of CH 2 in open-
chain compounds, thus:—
C 6 H ia 0*476, C 6 H J4 0*500, C 7 H 16 0*532, C 8 H]£ 0*554,
A 0*024 0*032 0*022
When considering an unknown value for a certain compound, it is
usually possible to find an analogous compound from which c may be
calculated, for example, cymene, C 10 H 14 , for the terpenes (menthane),
C 10 H 16 , methyl succinate for methyl maleate or fumerate, propionitrile
for ethel carbylamine, and ethyl nitroethane for ethyl nitrite.
Digitized by v^ooQle
APPENDIX II.
AN INVESTIGATION OF THE DICARBOXYLIC ESTERS.
Whilst this work was passing through the press, a successful investi¬
gation was made of the dicarboxylic esters, a series which for a long
time resisted treatment. They show quite a number of points of inter-
est, especially of the influence of complexity, so that it will be useful
to include them in an appendix at the end of the volume.
We find that the oxalates and formates show many points of
similarity, whilst the malonates and the succinates compare with the
acetates.
From the formates.
-COOCH,
62*7 - 4*0 as 587
- COOC 2 H 5
84*6 — 4*0 = 8o*6
Dimethyl oxalate.
Vm . (COOCH,) a
« 2 x 587 = 117*4
Observed 117*4
Methyl ethyl oxalate.
COOCH,
Vm |
COOC a H,
= 58*7 + 8o*6 = 139*3
Observed 139*4
From the acetates.
-COOCH,
83*2 - 25*9 = 57*3
-CH a . COOCH,
83*2 - 4*0 = 79*2
Dimethyl malonate.
COOCH,
V m CH-/
COOCH,
- 79 ' 2 + 587 - I 37‘9
Observed 137*9
The Oxalates (Wiens Konigsberg Inaug. Diss., 1887).
COOCH,
V w .
5nVa.
For complexity
+ » x 3*3
Calc.
toOCH,
COOCH,
”7*4
”7*4
—
117*4
toOC,H,
COOC.H,
1
139*4
139*4
—
139*4
COOC a H 6
cooc,h 7
162*2
161*8
—
161*8
toOC,H 7
COOC.H,
215*8
205*4
117*4 + 4 X 22*0
3 x 3*3
9*9
2i5*3
1
COOC 7 Hu
316-5
293*4
265
7 x 3*3
23*1
316-5
Digitized by v^ooQle
266
LIQUID CHEMICAL COMPOUNDS.
COOCjHj
toOC,H„
341*9 315*4
The Malonates.
/
COOCH,
CHo
\
COOCH,
COOC.H,
137*9
/
CH,
\
J n 5
/
CH,
\
COOC,H 5 1856
cooc,h 7
137*9
181*9
137*9 + 44 *o
COOC,H 7 235*0 225*9
The Succinates.
8 x 33
26*4
1 X 3*3
3 x 3*3
9*9
341-8
137*9
l85*2
235*8
CH,—COOCH,
llH, — COOCH,
CH,—COOC.Hj
1
160*0
CH,—COOC,H 7
CH,—COOC,H 5
257*2
248*0
160 + 4 X 22*0
3 x 3*3
9*9
257*9
im,—COOC,H,
CH,—COOC,H 5
2320
226*0
2 x 3*3
6*6
232*6
I:h,—cooc,h u
CH,—COOC 7 H 15
333-6
314*0
6 x 3*3
ig*8
333-8
Lh,—cooc 7 h 15
460*6
424*0
11 x 3*3
36*3
460-3
The rule for increase due to complexity in the case of Wien’s com¬
pounds is : Add 22*0 for every CH 2 added after the simple methyl com -
pounds , and an additional 3*3 for every CH 2 added beyond 3CH 2 . One
observation of some significance is, that the increase due to complexity
is rectilinear. This has been found to be the case in the
{ Acrylic esters CH, = CH—COOR
Phenolic ethers C 6 H 8 —OR and other aromatic esters
Dicarboxylic esters (COOR),, CH, (COOR),, C,H 4 (COOR),.
On the other hand we find that in the case of the aliphatic ethers
R—O—R 1 the augmentation depends upon the square of the complexity.
These effects are probably due to the following features. In the
Digitized by v^ooQle
APPENDIX II.
267 '
first series, the augmentation only affects the hydrocarbon (paraffinoid)
radicals R, and since the saturated carbon chain extends only on one
side, the augmentation depends on the complexity simply.
H—COOCH a . CH a . CH S . CH a . CH 8 .
In the aliphatic ethers the augmentation depends on the square of
the complexity , because the saturated carbon chain extends on both sides
of the oxygen atom.
<- - >
CH 8 . CH a . CH 3 . CH 3 —O—CH a . CH a . CH a . CH 8 .
There are indications that this is also the case in the symmetrical
normal paraffins.
CH S . CH a . CH a . CH a . CH a . CH a . CH a . CH S .
If the radicals on one side are substituted, or if the phenyl or other
similar radical replace the saturated aliphatic radical, so rendering the
compound non-symmetrical, the simple rule obtains. If, however, there
be a symmetrical arrangement on both sides the rule is more complex.
This more complicated rule suggests interaction between the two radicals .
This might be the case if the arrangement of the atoms were as in the
scheme.
CH S . CH a . CH a . CH 2
\
o
/
CH 3 .CH a .CH 2 .CH 3
The facts in favour of the disposition of the oxygen valencies shown
by ^>0 are numerous. Considerable differences in the complexity of
the two radicals generally cause a diminution however. This suggests
important modifications in the configuration of oxygen compounds or
rather they show that the configurations follow from a similar law to that
which governs the tetrahedral arrangement of groups about the carbon
atom.
ch 8 ch 8 ch 8 ch 8 .ch 2 ch 3 —ch 2 —c = o
Ah— ch c! = ch ch,— A
Ah, Ah, Ah, Ah, ch,^
In all these cases there are augmentations in the boiling-point and
diminutions in the volume.
It is noticed that among Wiens 1 compounds is the di-heptyl succin¬
ate, which contains no fewer than 53 atoms and the very large molecular
volume of 460 c.c. or nearly 500 c.c. This is the largest molecule with
which we have had to deal in the course of this investigation'.
Digitized by
Google
268
LIQUID CHEMICAL COMPOUNDS .
H
\
(
/
H
H
\
<
/
H
Dipeptyl Succinate C 9 H 4 (COOGyH^),.
O HHHHHHH
ijiujjj.
1111 jj i
H H H H H H H
HHHHHHH
-C—O— J— 0—1— J—J-J-J— H
!! I II I I II
O HHHHHHH
Finally it should be stated that some of Wiens' numbers differ un¬
accountably from the expected values.
COOC,H,
/
V*.
InVa.
A.
CH '\
COOC,H,
COOC 4 H*
/
208*2
210*5 *
— 2*3
CH*
\
cooc 4 h*
CH*—COOC.H,
280*0
286*4
— 6*4
<!h,—COOC 4 H,
278-4
283*2
— 4*8
it is at present impossible to account for these differences. They
are far larger than any probable error due to faulty experiment or even
to impurity of liquid. Another very remarkable feature is that Weger
gives data which are somewhat different to those which might be expected
from Wiens' results, and yet show a regularity so far as the increase due
to complexity is concerned. The compounds are succinates.
V m .
2*V a .
For complex calc. (Wiens).
CH 9 —COOCHj
160*0
<!h*—COOCH,
(Weger)
CH*—COOCH,
<!h*—cooc*h 4
185*0
182*0
1 x 3*3
185-3
182*0
(Weger)
A 33
CH*—COOCjH,
210*6
(Ih,—COOCjHj
210*3
204*0
2 x 3*3
207*3
(Weger)
A 3*3
CH*—COOC,H,
(!h,—COOC.H,
257*2
248*0
3 x 3*3
261*2
257*9
(Wiens)
A 3*3
* This is the volume of the isomeric ethyl succinate C 9 H 4 (COOC 9 H B ) 9 accord¬
ing to Weger.
Digitized by v^ooQle
APPENDIX II
269
In the first three cases Wiens* results are 3*3 units smaller than those
of Weger, but in the fourth case which was determined by Wiens, the
latter regularity applies and Weger’s regularity shows a result 3 *3 too
high. These facts, taken together with the anomalous results due to
Wiens which show differences equal to 3*3 or some multiple thereof,
leave us with the impression that these augmentations for some reason
unknown may in part fail. The principle involved is that a liquid sub¬
stance, which so far as we know, is the same structurally, may possess
two or more values, and that thus the volume of liquids may be variable
for unknown reasons. A similar case, is when a substance assumes the
liquid state, whereas the temperature conditions demand that it should be
a solid. The condition is, however, an unstable one, and slight causes
will lead to ordinary conditions. Another case is that of a liquid which
shows a variable surface tension owing to slight amounts of impurity.
In the light of this, we might suppose that the intermolecular forces are
also capable of being rendered latent and inoperative for a similar
reason.
A comparison of the boiling-points show considerable irregularities.
This might be expected to be the case.
The regularity demanded by Weger’s results nearly all show an aug¬
mentation after the methyl compound in the case of the esters. Wiens*
results all show an augmentation after the third carbon atom in the
radicals. There is thus a constant difference of about 3*3 between
Weger's results and those due to Wiens.
The Volumes of Geometrical Isomers .
It will now be interesting to calculate the volumes of a number of
geometrical isomers, constructed according to the well-known malenoid
and fumaroid types.
CH—X X—CH
II II
CH—Y CH—Y
“Adjacent.** “Opposed.**
In order to do this, we need to know the values of c for the corre¬
sponding saturated compounds. The data for the dicarboxylic esters
give us the necessary means of doing this and so of solving the problem.
The compounds open to investigation are :—
“ Adjacent.*’
Ethyl maleate.
C 8 H ia 0 4 b.p. 225-0
d* 1-06917
Propyl maleate.
b.p.
djo 1*02899
Methyl maleate.
CjHgCV^b.p. 205
dy 1-15172
“ Opposed.**
Ethyl fumarate.
b.p. 206*2
d«o 1*05200
Propyl fumarate.
d*o 1-02203
Digitized by v^ooQle
270
LIQUID CHEMICAL COMPOUNDS.
Methyl succinate.
B.p. d 0 .
d C.
B.p.
0*912 0*59
c,h X() o 4
195*2 1*1162 (<f 18 )
Ethyl succinate.
c 8 h 14 o 4
215*4 1*0596
0*82726 0*635
Ethyl propyl succinate.
c,h..o 4
231*0 1*03866
0*81476 0*658
Ethyl butyl succinate.
CioH] 8 0 4
247 1*02178
0*78572 0*639
Applying these values we find :—
Methyl succinate.
Methyl maleate.
CH a —COOCH 8
CH—COOCH 8
I
A
II
CH 2 —COOCH s
CH—COOCH 8
b.p. 195*2° V*» 160*0
-6*5
153*5 b.p. 205°
Ethyl succinate.
Ethyl maleate.
CH a —C00C 2 H 5
CH—COOC 2 H b
||
in.— cooc,h.
CH—COOC t H B
b.p. 2x5® V)» 210*3
— 7 *o
203*3 b.p. 225°
Propyl succinate.
Propyl maleate.
CH,—COOCjHj
CH—COOC 8 H 7
||
Ah,—COOC,H 7
CH—COOC 8 H 7
b.p. 247 0 V m 258*4
—8*4
250*0 b.p. 257°
The Fumarates.
Ethyl fumarate C 8 H 12 0 4 .
C 2 H B OOC—CH
II
HC—COOC 2 H B
b.p. 218° d % o 1*05200
Vw 204*5
A + 1*2
Ethyl maleate <203*3
Propyl fumarate ^ 10 ^ 16 ^ 4 *
C 3 H 7 . OOC—CH
II
HC—COOC 8 H 7
b.p. 250° d % 01*02203
V*» 250*8
A + o*8
Propyl maleate 250*0
It is seen that the maleic esters differ from the corresponding succinic
esters by about the volume of H 2 = 7 *2 which represents their difference
in composition.
The furmaric esters are about a unit larger.
We might expect the former to manifest the above differences. At
any rate it shows that the succinic and no doubt the malonic and oxalic
esters are constructed according to the adjacent types.
COOR COOR CH 2 —COOR
I CH, |
COOR 1
\
COOR 1
CH„—COOR 1
Digitized by v^ooQle
APPENDIX II
271
The smallness of the differences between the volumes of members
of the two series is doubtless owing to the small reactivity of the groups
COOR. This is in conformity with chemical observation which shows
that succinic anhydride may be formed from succinic derivatives.
If we try to account for the actual volumes of the succinates and the
maleates, etc., we obtain the following results :—
Ethyl formate HCOOC 2 H 6 Vm 84*6
Ethyl propionate CH 3 . CH 2 . COOC 2 H B 1277
CH a —COOCoH s
c 2 h 4
84*6 - 4*0 = 8o*6
2fcOOC 2 H B 161 *2
43 *i
2CH 2 43*1
CH 2 —COOC 2 H b
For complexity
204*3
+ 6*6
'ZnVa 210*9
Ethyl succinate Vm 210*3
2COOC 2 H b 161*2
CH—COOC 2 H b
II zCH 35-9
CH—COOC,H 6 197-1
For complexity + 6*6
2»V a 203*7
Ethyl maleate V m 203*3
The above instances of geometrical isomerism are too few in number
upon which to build any general hypothesis, they merely serve as indi¬
cations of what might be expected.
Digitized by v^ooQle
Digitized by
REFERENCES.
I Le Bas, J.C.S., Trans., 91, 112 (1907).
9 Krafft, Ber., 15, 1687, 1711 (1882); 19, 2218 (1886).
3 Le Bas, Phil. Mag. [6], 14, 81, 333 (1907).
4 Baeyer, Ber., 10, 1286-7 ( i8 77 )*
5 Le Bas, Journ. Chem. Soc. Proc., 27, 196 (1911).
6 Biach, Zeits. Phys. Chem., 50, 43 (1905).
7 Le Bas, Phil. Mag. [6], 16, 91, 66 (1908).
8 Krafft, Ber., 16, 3018 (1883); 17, 1371 (1884).
9 Krafft, Ber., 15, 1728 (1882).
10 Schiff, Ann., 223, 247 (1884).
II Kopp, Ann., 96, 153, 303 (1855).
12 Le Bas, Phil. Mag. [6], 16, 91, 60 (1908).
18 Buff, Ann. Suppl., 4, 129 (1865).
34 Schiff, Ann., 220, 301 (1883).
15 Lossen, Ann., 254, 42 (1889).
16 Gartenmeister, Ann., 233, 249 (1886).
17 Dobriner, Ann., 243, 11 (1888).
18 Schiff, G.C.I., 13, 177 (1883).
19 Pinette, Ann., 243, 32 (1888).
20 Zander, Ann., 225,174 (1884).
21 Young, Brit. Assoc. Report, Cambridge (1904).
22 Le Bas, Phil. Mag. [6], 14, 81, 337 (1907).
23 Naumann, Ber., 7, 173, 206 (1874).
24 Schiff, Ber., 15, 2974 (1882).
25 Pinette, Ann., 243, 50 (1888).
26 Neubeck, Zeits. Phys. Ch., 1, 656 (1887).
27 Perkin, Journ. Chem. Soc., Trans., 81, 292 (1902).
28 (1) Kehrmann, Journ. Chem. Soc., Trans. ?
(2) O. M. Foster, Journ. Chem. Soc., Trans., 81, 268 (1902).
29 Schiff, Gazz. C. Ital., 13, 177 (1881).
30 Le Bas, Phil. Mag. [6], 28, 439 (1914).
31 F. Swartz, Journ. Chem. Soc., Abs. i., 129, 82 (1901).
•— 32 Thorpe, Journ. Chem. Soc., Trans., 37, 201, 371 (1880).
38 Jungffeisch, Jahresb. liber die Fortschritte der Chemie, 19, 551; 20, 36.
34 Le Bas, Phil. Mag., 27, 988 (1914).
35 Le Bas, Chem. News (1909).
86 Le Bas, Phil. Mag., 27, 976 (1914).
^. 37 Thorpe, Journ. Chem. Soc., 37, 360 (1882).
as Weger, Ann., 221, 61 (1883).
89 Dobriner, Ann., 243, 1 (1888).
40 Le Bas, Phil. Mag. [6], 27, 750 (1914).
41 Gartenmeister, Beibl., 9, 766 (1885); Ann., 233, 249 (1886).
42 Hinrichs, Jahresb., etc., 1868, p. 80.
43 Pinette, Ann., 243, 32 (1888).
273
Digitized by v^ooQle
*74
Liquid chemical compounds .
44 Schiff, G.C.I., 13, 177 (1883).
44 Zander, Ann., 224, 48 (1884); Schiff, G.C.I., 13, 177 (1883).
48 Gartenmeister, Ann., 233, 249 (1886); Schiff, G.C.I., 13, 177 (1883).
Els&sser, Ann., 2x8, 302 (1883).
47 Wiens, Konigsberg, Inaug. Diss. (1887); Weger, Ann., 221, 61 (1883).
48 Weger, Ann., 221, 61 (1883).
44 Schiff, G.C.I., 13, 177 (1883).
#0 Weger, Ann., 221, 61 (1883).
41 Weger, do., do., do.
48 Ramsay, Journ. Chem. Soc., Trans., 35, 471 (1879).
48 Schiff, Ber., 18, 1605 (1885).
44 Briihl, Ber., 39, 2752 (1906); 40, 4183 (1907).
Digitized by v^ooQle
BIBLIOGRAPHY.
Kopp, Pogg. Ann., 47, 113 (1839); 5 *» 243 , 262 (1844); Ann., 41, 79 (1842); 96,
153 , 303 (1855); 250 , 1 (1889).
Constitutive Relations.
(z) The Homologous Increment :—
Gartenmeister, Ann., 233, 249 (1886); Pinette, Ann., 243, 32 (1888); Dobriner,
Ann., 243, n (1888); Elsasser, 218, 337(1883); Zander, Ann., 225, 74
(1884); Lossen, 254, 42 (1889).
(2) Isomerides: —
Horstmann, Beziehungen zwischen der Raumerfullung und chemischer
Zusammensetzung; Graham-Otto Lehrbuch der Chemie, vol. i., 3 (1898).
Krafft, Ber., 15, 1711 (1882); 17, 1371 (1884).
Brown, Proc. Roy. Soc., 26, 247 (1878); Dobriner, Ann., 243, 30 (1888):
Fettler, Zeit. Phys. Chem., 4, 66 (1889).
Gartenmeister, Ann., 233, 249 (1886); Neubeck, Zeit. Phys. Chem., I, 649
(1887); Pinette, Ann., 243 (1888); Ramsay, Trans. Chem. Soc., 35, 463
(1879); Schiff, Ann., 220, 71, 278 (1883); Ber., 14, 2761 (1881); Stadel,
Ber., IS 2559 (1889).
Thorpe, Trans. Chem. Soc., 37, 141, 327 (1880); Zander, Ann., 224, 74
(1884); Schroder, Ber., 13, 1560 (1880).
Stereoisomerides :—
Walden, Ber., 29, 1699 (1896); Liebisch, Ann., 286, 140 (1895); Traube,
Ann., 240, 43 (1886).
Unsaturation :—
Buff, Ann. Suppl., 4, 129 (1865); Schiff, Ann., 220, 301 (1883); Lossen,
Ann., 254, 42 (1889); 214, 81 (1883); Krafft, Ber., 17, 1371 (1884);
Zander, Ann., 214, 138 (1882); Horstmann, Graham-Otto Lehrbuch
der Chemie, vol. i., 3, p. 422 (1880); Schiff, Ann., 220, 71 (1883);
Zander, Ann., 114, 138 (i860); Krafft, Ber., 17, 1371 (1884).
Ring formation :—
Willstatter, Ber., 41, 1480 (1908).
The author’s work deals with all of the above features.
The chief papers are:—
Trans. Chem. Soc., 91,112 (1907); Phil. Mag. [6], 14,324 (1907); 16, 60 (1908);
Chem. News, 98, 85 (1908); 99, 206 (1909); Phil. Mag., 27, pp. 344, 741, 976; 28,
P- 439 (1914)-
The Influence of Constitution on the Molecular Volumes of Organic Com¬
pounds at the Boiling-point. Brit. Assoc. Meeting, Section B, Portsmouth
(1911).
The Influence of the Alternating Factor in certain series on the Molecular
Volumes at the Melting-point. Proc. Chem. Soc., 27, 196 (1911).
A Formula by means of which the Molecular Volume at the Boiling-point may
be Calculated. Proc. Chem. Soc., 30, 86 (1914).
Molecular Volume Theories and their relations to current conceptions of
Liquid Structure Science Progress, 32, 663 (1914).
275
Digitized by v^ooQle
PRINTED IN GREAT BRITAIN BY THE UNIVERSITY PRESS, ABERDEEN
Digitized by v^ooQle
Digitized by v^ooQle
RETURN TO the circulation desk of any
University of California Library
or to the
NORTHERN REGIONAL LIBRARY FACILITY
University of California
Richmond Field Station, Bldg. 400
1301 South 46th Street
Richmond, CA 94804-4698
ALL BOOKS MAY BE RECALLED AFTER 7 DAYS
To renew or recharge your library materials, you may
contact NRLF 4 days prior to due date at (510) 642-6233
DUE AS STAMPED BELOW
JAN 16 2008
Jl
DD20 12M 7-06
Digitized by
Google
I
YC 60621
\lA
>* I* U'
( 404 - 0 / ;
TO fr: ' 1
UNIVERSITY OF CALIFORNIA LIBRARY
Digitized by
Google
